Capacity Mechanisms in the EU Energy Market: Law, Policy, and Economics 9780198749257, 9780191066177

Citation preview

CAPACITY MECHANISMS IN THE EU E N E R G Y MA R K E T

Capacity Mechanisms in the EU Energy Market Law, Policy, and Economics

Edited by LEIGH H ANCHER ADRIEN DE HAUTECLOCQUE and MAŁ GORZATA S ADOWSKA

1

3 Great Clarendon Street, Oxford, OX2 6DP, United Kingdom Oxford University Press is a department of the University of Oxford. It furthers the University’s objective of excellence in research, scholarship, and education by publishing worldwide. Oxford is a registered trade mark of Oxford University Press in the UK and in certain other countries # The several contributors 2015 The moral rights of the authors have been asserted First Edition published in 2015 Impression: 1 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, without the prior permission in writing of Oxford University Press, or as expressly permitted by law, by licence or under terms agreed with the appropriate reprographics rights organization. Enquiries concerning reproduction outside the scope of the above should be sent to the Rights Department, Oxford University Press, at the address above You must not circulate this work in any other form and you must impose this same condition on any acquirer Crown copyright material is reproduced under Class Licence Number C01P0000148 with the permission of OPSI and the Queen’s Printer for Scotland Published in the United States of America by Oxford University Press 198 Madison Avenue, New York, NY 10016, United States of America British Library Cataloguing in Publication Data Data available Library of Congress Control Number: 2015944364 ISBN 978–0–19–874925–7 Printed and bound by CPI Group (UK) Ltd, Croydon, CR0 4YY Links to third party websites are provided by Oxford in good faith and for information only. Oxford disclaims any responsibility for the materials contained in any third party website referenced in this work.

Editors’ preface How to ensure adequate levels of generation capacity in the newly liberalized energy markets? Twenty years into energy market liberalization, Member States of the European Union (EU) start to question the ability of the so-called ‘energy-only’ markets, where generators are paid only for the energy they produce, to provide appropriate incentives to build new generating capacity in the right quantity, the right location, and based on the right technology. Why is there so little faith in market forces? Is not an optimal level of generation investments something that a liberalized and wellfunctioning market should provide? A reply frequently heard from economists is that, in reality, truly free and competitive energy-only markets do not exist. A number of political/regulatory constraints, such as price caps, or operational barriers, keep prices for wholesale and balancing energy below their efficient levels at times when they should be high, providing insufficient revenues for gas-fired peaking units to recover their capital costs. The increasing intake of subsidized renewable energy into the system likewise depresses wholesale prices and drives higher-cost conventional plants out of business. However, as suggested by the diverging views presented in this book, there is still no consensus as to whether these concerns are valid or not. And the common cure envisaged these days? Even more subsidies. We are witness to a hasty and uncoordinated introduction of so-called capacity remuneration mechanisms (or simply, capacity mechanisms) in a number of European countries. In essence, capacity mechanisms are just another form of state-driven support which compensates generators for their capacity, that is, their availability to produce energy at some point in the future. A more certain and stable stream of revenue for capacity, in addition to revenue from the sale of energy, is supposed to mitigate generators’ investment risk, encouraging them to build new power plants, and also to keep the existing ones in operation. Capacity mechanisms can also remunerate consumers for their commitment to reduce energy consumption at some point in the future. In this case, the mechanism supports investments in demand side response solutions. Capacity mechanisms are thus nothing more than a new regulatory ‘patch’ in the long transition towards a competitive, sustainable, and secure energy market the design of which still remains to be perfected. But is not the cure worse than the disease—assuming there indeed is a disease? Capacity mechanisms may have serious implications for the completion of the European internal electricity market, the Holy Grail pursued since the mid-1990s. In particular, they might hamper cross-border trade and distort competition in the national day-ahead and balancing markets. They might also distort investment signals in the internal market leading to locational over- and/or under-capacity. As the economic rationale for introducing capacity mechanisms still remains uncertain, their actual impact on the performance of markets is even harder to establish. How would their introduction affect competition in spot markets? How would cross-border trade evolve? Would capacity mechanisms really boost new investments, as opposed to maintaining old and inefficient power plants on a financial drip? Would they optimize

vi

Editors’ preface

the generation mix? Would they incentivize demand side response and electricity storage? These are only some of the questions facing European and national policy makers and regulators. On 29 April 2015, the European Commission (the Commission) launched a Europewide sector inquiry into capacity mechanisms, the first and one of a kind inquiry into State aid across an industry sector.1 The Commission aims to gain better understanding of capacity mechanisms, and in particular whether any specific schemes would distort the internal energy market. In this context, it is important first to make the right diagnosis and to try to identify cheaper and less distortive alternatives to just another form of support scheme. Refining current market designs, in particular by fostering demand response, improving the operation of balancing markets and increasing interconnection might provide more suitable means to ensure security of supply, even if these types of instruments all carry their own difficulties and complexities. From the perspective of the internal energy market what is required is to strike a careful balance between the efficacy of the national measure and its distortive effects on competition and cross-border trade. Achieving such a balance is one of the most analytically and politically challenging tasks for national and European regulators alike. If we assume that capacity mechanisms are indeed necessary to achieve an adequate level of investment, their design then becomes the core question. As discussed at length in this book, there is a wide variety of capacity mechanisms to choose from, from simple payments for capacity to more complex capacity auctions. Capacity mechanism design raises a classical trade-off between a centralized, administrative determination of prices and/or quantities, and leaving market forces to reign as freely as possible. Rigid designs with a higher level of public intervention will probably ensure realization of the target level of investment, but at what cost? Market-based designs, although superior in theory, may at the same time leave scope for anticompetitive behaviour, and may not bring the expected outcomes. Another important question is whether a given capacity mechanism could fit well into the EU Target Model for electricity markets. One must remember that capacity mechanisms add a layer of regulatory complexity onto pre-existing market dynamics when all market design elements are closely related and interdependent. A mismatch between capacity mechanisms and existing or new market elements is probably unavoidable, requiring continuous monitoring and adaptation ex post by regulators. From the perspective of European electricity market integration, the participation of cross-border capacity remains a key element of any capacity mechanism design. Member States implement capacity mechanisms for very different reasons, and tailor them to their national markets. The resulting variety of capacity mechanism calls into question the feasibility of their eventual harmonization at the EU level. However, as many of the contributors to this volume maintain, some degree of harmonization (for instance, at regional level) would significantly mitigate the very real risks of distortions in the internal market. It is therefore not surprising that the Commission closely 1

Press release IP/15/4891 of 29 April 2015. More information on the sector inquiry is available at http:// ec.europa.eu/competition/sectors/energy/state_aid_to_secure_electricity_supply_en.html, accessed 23 July 2015.

Editors’ preface

vii

investigates the issue, and has developed a set of criteria which Member States should follow when designing capacity mechanisms. While the Commission cannot prevent Member States from implementing capacity mechanisms to address genuine generation adequacy issues, the Treaty does confer certain albeit limited powers to allow the Commission to control these national developments. As discussed in depth in this book, the Treaty provisions on state aid, public services, competition, and the free movement of goods all have useful and important roles to play. Overall, concerns about security of supply in Europe, and plans to introduce capacity mechanisms, raise more questions than they answer. Drawing on policy, economic and legal expertise, this book provides a comprehensive study on capacity mechanisms, their causes and potential consequences for the internal electricity market and finally, and the role of EU law in mitigating the risk of distortion of inter-state trade. This book project originated from a comparison of eleven in-depth case studies of national capacity mechanisms existing and/or planned in Europe. The findings extracted from these various country models were presented on 11 October 2013 at a Florence School of Regulation (FSR) workshop in Florence, and provided a valuable research base for the chapters dealing with the policy, economics, and law of capacity mechanisms from a broader European and international perspective, included in the first three parts of this book. National case studies form the fourth part of this volume. Part I deals with EU policy making. Chapter 1, written by Francisco Enrique GonzálezDíaz, provides an introduction to the book. It explains what capacity mechanisms are and why they are so much discussed at the European level. The chapter guides the reader through the different types of capacity mechanisms, and discusses relevant EU legal measures, policy documents, and individual cases, placing capacity mechanisms in the broader EU legal and regulatory framework. In chapter 2, Alberto Pototschnig and Martin Godfried contend that the lack of cross-border coordination of national generation adequacy policies is to the detriment of the market integration process. Representing the view of the Agency for the Cooperation of Energy Regulators (ACER), the authors suggest that enhanced coordination of these policies would minimize the risk of market distortion, in particular by developing common criteria for the assessment of generation adequacy. This recommendation is taken further by Jean-Michel Glachant and Arthur Henriot who, in chapter 3, propose a set of conditions for coordinating generation adequacy policies at EU level. The authors’ analysis takes into account the specificities of European electricity markets: progressive integration and an increasing share of renewable energy. Part II brings together various economic insights on capacity mechanisms. It begins with chapter 4 by Christoph Riechmann and Jens Perner who get to the root of the problem: can energy-only markets with rapid expansion of renewable energy deliver adequate levels of investment? If not, can we ensure investments without resorting to capacity mechanisms? The authors identify different measures which may still fall within the current energy-only market design as opposed to those which may go beyond that market design. They suggest that reform of the existing market design should be exploited first before embarking on the introduction of more extensive mechanisms in the form of capacity mechanisms. However, as they acknowledge, divergent security of supply problems may require different measures. This latter issue is further developed in chapter 5, in which Fabien Roques and Charles Verhaeghe

viii

Editors’ preface

take a closer look at the specific local drivers for the implementation of capacity mechanisms in various Member States, and return to the issue of a more coordinated approach to capacity mechanisms at the European level, in particular in terms of crossborder participation. The feasibility of cross-border participation in capacity mechanisms is thoroughly studied by Dominique Finon in chapter 6. The author distinguishes two forms of cross-border participation: (a) explicit, with cross-border trade in capacity rights, and (b) implicit, which is simply based on the statistical contribution of generation capacity located in one country to the system reliability of another country. He examines the social benefits of these two forms of cross-border participation in situations with and without cross-border congestion. Chapter 7 looks beyond the EU debate on capacity mechanisms and draws on broader international experience. Carlos Batlle, Paolo Mastropietro, Pablo Rodilla, and José Ignacio Pérez-Arriaga analyse various design elements of selected American long-term auctioning mechanisms, as a basis for recommendations to support policy developments currently underway in the European Union context. Part II closes with chapter 8 by Bert Willems, who provides a more theoretical economic analysis of a capacity mechanism in a simplified perfectly competitive electricity market. The author shows how a technology-neutral capacity mechanism can restore efficiency caused by a detrimental effect of a price cap on shortterm profitability and long-term investment incentives. Part III focuses on the role of EU law in mitigating the risks that a patchwork of different national capacity mechanisms can pose to the successful functioning of the internal energy market. The EU state aid rules are currently the primary instrument at the Commission’s disposal when it comes to controlling capacity mechanisms, as explained by Leigh Hancher in chapter 9. The author discusses circumstances in which a national capacity mechanism can constitute state aid, and considers how farreaching the criteria developed by the Commission in its recent soft-law state aid guidelines are likely to be in practice in preventing distortions of the internal energy market. While state aid rules can be applied to vet the design of capacity mechanism ex ante and therefore ideally at the implementation stage, chapter 10 turns to the EU antitrust rules, which are applied ex post. Adrien de Hauteclocque and Małgorzata Sadowska look at the potential for different types of anticompetitive behaviour by firms in the context of capacity mechanisms. They argue that in certain cases antitrust rules might not provide an effective response. This highlights the importance of regulation and market monitoring in mitigating the risk of market abuse. Finally, national capacity mechanisms can entail barriers to cross-border trade in energy, a result which appears to be hard to reconcile with the free movement of goods, one of the fundamental freedoms enshrined in the EU Treaties. In chapter 11, Peter Oliver discusses recent landmark cases on the compatibility of green certificate schemes in Sweden and Belgium, with the Treaty fundamental freedoms and their significance for the capacity mechanism debate. Part IV contains detailed case studies of capacity mechanisms existing or planned in eleven Member States, most of which are now subject to the Commission’s sector inquiry. All chapters follow the same structure to facilitate comparison between individual country models, and each chapter is divided into three sections. The first section gives a brief description of the market context, highlighting relevant national

Editors’ preface

ix

regulation on security of supply and assessment of generation adequacy. The second section focuses on capacity mechanism design and, in case of existing mechanisms, looks at their functioning in practice. The third and last section adds a European dimension to the discussion, briefly assessing the mechanisms’ compatibility with EU law. This book is to provide a first point of reference for regulators and policy makers as well as energy firms who are either involved in the design of capacity mechanisms in Europe or are dealing with newly introduced schemes. We hope that the various contributions to this book will promote a deeper understanding of how these mechanisms work and can stimulate wider debate on their implications both for national markets, as well as the European internal market. We hope this book will prove an invaluable resource for anyone interested in energy market design, regulation, and competition issues. We would like to take this opportunity to thank all the people who made this project so rewarding for us. We are deeply grateful to each of them. This book would not have been realized without the continuous support of the Florence School of Regulation at the European University Institute. The FSR is a unique organization in the EU, at the crossroads of academia and policy making. We owe a special thanks to its director (and a contributor to this book), Professor Jean-Michel Glachant for his encouragement and unwavering support, as well as to our sponsors who make the development of such projects possible. We would like to thank all of the contributing authors for their enthusiasm and dedication to the project, and for their patience with the demands of an extensive editorial process. Every chapter was anonymously revised and crossreviewed by fellow contributors. We address our sincere thanks to each author for having participated in this endeavour. In addition to the contributors, we would like to thank Anne Marie Kehoe and Claire Local who provided invaluable assistance in proofreading, editing, and preparing the manuscript for publication. At Oxford University Press, Alex Flach, Natasha Flemming, Kizzy Taylor-Richelieu, and Emma Slaughter were inspiring and patient partners. Leigh Hancher Adrien de Hauteclocque Małgorzata Sadowska

Table of Contents Table of Cases Table of Legislation Table of Relevant Non-Legislative Documents List of Contributors List of Abbreviations

xv xix xxvii xxxi xxxix

PART I. POLICY 1. EU Policy on Capacity Mechanisms 1.1 1.2 1.3 1.4

Why capacity mechanisms? The missing money problem and RES capacity Core features of capacity mechanisms The EU approach to capacity mechanisms before the 20/20/20 Package The EU approach to capacity mechanisms after the 20/20/20 Package (and before the November 2013 Communication) 1.5 The November 2013 Communication 1.6 Conclusion—where next with capacity mechanisms in the EU?

2. The Regulators’ View: ACER’s Report on Capacity Mechanisms and the EU Internal Electricity Market 2.1 2.2 2.3 2.4 2.5

Introduction The contribution of energy-only markets to generation adequacy Impact of capacity mechanisms: design and distortions Cross-border participation in capacity mechanisms Conclusions and recommendations

3. Capacity Mechanisms in the European Market: Now, but How? 3.1 3.2 3.3 3.4 3.5

Introduction The pervasive impact of capacity mechanisms on the remuneration of flexibility The costs of self-sufficient capacity mechanisms A framework for coordination of national capacity mechanisms at EU level Conclusion

3 3 7 16 18 22 30 32 32 33 35 37 38 40 40 41 44 48 54

PART II. ECONOMICS 4. Energy Market Design with Capacity Mechanisms 4.1 4.2 4.3 4.4 4.5 4.6

Introduction Energy-only market—can it be sustainable? What are the policy options—and how would they perform? How do renewable energies fit into the market design? What are the challenges in the EU context? Summary—when to use which capacity mechanisms?

5. Different Approaches for Capacity Mechanisms in Europe: Rationale and Potential for Coordination? 5.1 Introduction 5.2 A patchwork of capacity mechanisms in Europe: different designs fit different needs

59 59 60 67 72 75 76 79 79 80

xii

Table of Contents 5.3 5.4 5.5 5.6

Toward coordination of capacity mechanisms in Europe Capacity mechanisms and the new EEAG 2014–2020: Key issues Coordinating capacity mechanisms in Europe Conclusions

6. Capacity Mechanisms and Cross-Border Participation: The EU Integrated Approach in Question 6.1 6.2 6.3 6.4 6.5

Elements of the EU debate on cross-border participation Explicit cross-border participation with congested capacity of interconnections Social efficiency of cross-border participation from the national perspective Social efficiency of cross-border participation from the EU perspective Conclusion

7. The System Adequacy Problem: Lessons Learned from the American Continent 7.1 7.2 7.3 7.4 7.5 7.6 7.7 7.8

Introduction Design elements Target market Lag period (or lead time) Contract duration Defining the requirements associated with the ‘reliability product’ Indexation and warranties Summary and high-level recommendations

8. The Generation Mix, Price Caps, and Capacity Payments 8.1 8.2 8.3 8.4 8.5 8.6 8.7

Introduction Optimal generation mix A competitive market leads to an optimal generation mix A price cap distorts investment levels Capacity payments restore efficiency Renewable energy and demand participation Conclusion

81 86 89 93 95 95 102 106 110 117 119 119 121 121 127 128 130 136 138 140 140 141 145 147 149 150 153

P A RT III. LAW 9. Capacity Mechanisms and State Aid Control: A European Solution to the ‘Missing Money’ Problem? 9.1 9.2 9.3 9.4 9.5 9.6

Introduction Policy evolution Funding capacity mechanisms: When do the state aid rules apply? Compatible support—the new EEAG 2014–2020 Assessment Conclusion

10. Antitrust Law: A Missing Piece in a Regulatory Puzzle? 10.1 10.2 10.3 10.4 10.5 10.6

Introduction EU antitrust enforcement in the energy sector—a primer A typology of antitrust issues related to capacity mechanisms The problem with the market definition Selected anticompetitive practices under Articles 101 and 102 TFEU Conclusions

157 157 158 162 171 175 181 182 182 183 185 187 189 200

Table of Contents

11. Free Movement of Goods in the Labyrinth of Energy Policy and Capacity Mechanisms 11.1 11.2 11.3 11.4 11.5 11.6

Introduction Energy as goods Free movement of goods and state aid Free movement of goods: Restrictions Free movement of goods: Justification Conclusion

xiii

201 201 202 203 203 210 223

PART IV. CASE S TUDIES 12. Austria

227

12.1 12.2 12.3 12.4 12.5

227 227 235 237 240

Introduction Setting the scene Energy-only market European dimension Conclusion

13. Belgium 13.1 13.2 13.3 13.4 13.5

Introduction Setting the scene Capacity mechanism European dimension Conclusion

14. France 14.1 14.2 14.3 14.4 14.5

Introduction Setting the scene Capacity mechanism European dimension Conclusion

15. Germany 15.1 15.2 15.3 15.4 15.5

Introduction Setting the scene Energy-only market and network reserve European dimension Conclusion

241 241 241 245 253 255 256 256 256 258 264 270 271 271 271 276 285 287

16. Greece

288

16.1 16.2 16.3 16.4 16.5

288 288 293 297 301

Introduction Setting the scene Capacity mechanism European dimension Conclusion

17. Italy 17.1 17.2 17.3 17.4 17.5

302 Introduction Setting the scene Capacity mechanism European dimension Conclusion

302 302 306 310 313

xiv

Table of Contents

18. Netherlands 18.1 18.2 18.3 18.4

Introduction Setting the scene Energy-only market and the European dimension Conclusion

314 314 314 318 320

19. Norway

321

19.1 19.2 19.3 19.4 19.5

321 321 325 331 334

Introduction Setting the scene Energy-only market European dimension Conclusion

20. Poland 20.1 20.2 20.3 20.4 20.5

Introduction Setting the scene Energy-only market, capacity measures, and a ‘capacity debate’ European dimension Conclusions

21. Spain 21.1 21.2 21.3 21.4 21.5

Introduction Setting the scene Capacity mechanism European dimension Conclusion

22. United Kingdom 22.1 22.2 22.3 22.4 22.5

Index

Introduction Setting the scene Capacity mechanism European dimension Conclusion

335 335 335 342 347 349 351 351 351 356 363 364 365 365 365 370 380 382 383

Table of Cases EU COURTS (alphabetical order) Adria-Wien Pipeline and Wietersdorfer & Peggauer Zementwerke (Case C-143/99) [2001] ECR I-8365 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .16.4.1 Air France v Commission (Case T-358/94) [1996] ECR II-2109. . . . . . . . . . . . . . . . . . . . . . . . . .9.3.4 Ålands Vindkraft AB v Energimyndigheten (Case C-573/12) 1 July 2014, not yet reported. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11.1, 11.4.4, 11.5.4, 11.5.4.2 Almelo v Energiebedrijf Ijsselmij (Case C-393/92) [1994] ECR I-1477. . . . . . . . . . . . . . . . . . . . . 11.2 Altmark Trans (Case C-280/00) [2003] ECR I-7747 . . . . . . . . . . . . . . . . . . . . . . . . 1.3.2, 9.3.3, 16.4.1 Association Vent de Colère! and others (Case C-262/12) 19 December 2013, not yet reported . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .9.3.4 Austria v Commission (Case T-251/11) 11 December 2014, not yet reported . . . . . . . . . . . . . . .9.3.4 Bauhuis (Case 46/76) [1977] ECR 5 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .11.4.2 Belgium v Commission (Maribel) (Case C-75/97) [1999] ECR I-3671 . . . . . . . . . . . . . . . . . . . .16.4.1 Belgium v Spain (Rioja) C-388/95 [2000] ECR I-3123 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .11.4.5 Bouhelier (Case 53/76) [1977] ECR 197. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .11.5.3 Bouygues and Bouygues Telecom v Commission (Joint Cases C-399/10 P and C-401/10 P) 19 March 2013, not yet reported . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .14.4.3 BUPA and others v Commission (Case T-289/03) [2008] ECR II-81 . . . . . . . . . . . . . . . . . . . . . . . 1.5 Campus Oil v Minister of Industry and Energy (Case 72/83) [1984] ECR 2727. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .11.4.4, 11.5.2.1, 11.5.4.1, 16.4.2 Co-Frutta (Case 193/85) [1987] ECR 2085 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .11.4.3 Commission v Austria (Case C-320/03) [2005] ECR I-9871 (Brenner I). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11.5.2.2, 11.5.2.3, 11.5.4 Commission v Belgium (Case 132/82) [1983] ECR 1649. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .11.4.2 Commission v Belgium (Case C-2/90) [1992] ECR I-4431 . . . . . . . . . . . . . . . . . . . . . . . . . . . 11.5.4.2 Commission v DEI (Case C-553/2012) 17 July 2014, not yet reported . . . . . . . . . . . . . . . . . . . .16.2.1 Commission v Denmark (returnable bottles) (Case 302/86) [1988] ECR 4607. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11.4.3, 11.5.4, 11.5.4.2 Commission v Denmark (vitamins) (Case C-192/01) [2003] ECR I-9693 . . . . . . . . . . . . . . . 11.5.2.1 Commission v France (Case C-159/94) [1997] ECR I-5815 . . . . . . . . . . . . . . . . . . . . . . . . . . 11.5.4.1 Commission v France (Case C-483/99) [2002] ECR I-4781 . . . . . . . . . . . . . . . . . . . . . . . . . . . .14.4.1 Commission v France (alcohol advertising) (Case 152/78) [1980] ECR 2299. . . . . . . . . . . . . . .11.5.3 Commission v France (homeopathic medicines) (Case C-212/03) [2005] ECR I-4213 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11.5.2.1 Commission v France (processing aids) (Case C-333/08) [2010] ECR I-757 . . . . . . . . . . . . . 11.5.2.1 Commission v Germany (Case 18/87) [1988] ECR 5427. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .11.4.2 Commission v Germany (Case T-134/14) appeal brought on 28 February 2014, pending . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .15.2.1 Commission v Germany (garlic capsules) (Case C-319/05) [2007] ECR I-9811 . . . . . . . . . . . 11.5.2.1 Commission v Germany (hospital pharmacies) (Case C-141/07) [2008] ECR I-6935 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11.5.2.1 Commission v Germany (packaging) (Cases C-463/01) [2004] ECR I-11705. . . . . . . . . . . . . 11.5.2.3 Commission v Germany (phytopharmaceuticals) (Case C-114/04) 14 July 2005, not yet reported . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11.5.2.3 Commission v Greece (Greek Oil Supplies I) (Case C-347/88) [1990] ECR I-4747 . . . . . . . . 11.5.4.1 Commission v Greece (Greek Oil Supplies II) (Case C-398/98) [2001] ECR I-7915 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11.5.4.1, 16.4.2 Commission v Ireland (Cyprus potatoes) (Case 288/83) [1985] ECR 1761 . . . . . . . . . . . . . . 11.5.4.1

xvi

Table of Cases

Commission v Ireland (souvenirs) (Case 113/80) [1982] ECR 1625 . . . . . . . . . . . . . . . . . . . . . .11.5.4 Commission v Italy (Case 7/68) [1968] ECR 423, 428. . . . . . . . . . . . . . . . . . . . . . . . . . . . 11.2, 11.4.2 Commission v Italy (electricity monopoly) (Case C-158/94) [1997] ECR I-5789 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11.2, 11.5.4.1 Commission v Italy (import deposits) (Case 95/81) [1982] ECR 2187 . . . . . . . . . . . . . . . . . . 11.5.4.1 Commission v Italy (pork) (Case 7/61) [1961] ECR 317 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11.5.4.1 Commission v Italy (statistical levies) (Case 24/68) [1969] ECR 193 . . . . . . . . . . . . . . . . . . . . .11.4.2 Commission v Italy (trailers) (Case C-110/05) [2009] ECR I-519. . . . . . . . . . . . . . . . . . . . . . 11.5.2.1 Commission v Netherlands (Case C-157/94) [1997] ECR I-5699 . . . . . . . . . . . . . . . . . . . . . . 11.5.4.1 Commission v Netherlands (C-279/08 P) [2011] ECR I-7671. . . . . . . . . . . . . . . . . . . . . 12.4.2, 14.4.3 Commission v Portugal (tinted film) (Case C-265/06) [2008] ECR I-2245. . . . . . . . . . . . . . . 11.5.2.4 Commission v Spain (Case C-160/94) [1997] ECR I-5851 . . . . . . . . . . . . . . . . . . . . . . . . . . . 11.5.4.1 Commission v Spain (Case C-207/07) [2008] ECR I-111 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .14.4.1 Compagnie Commerciale de l’Ouest (Joined Cases C-78/90 to C-83/90) [1992] ECR I-1847 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .9.3.4 Conegate v HM Customs and Excise Commissioners (Case 121/85) [1986] ECR 1007 . . . . . . .11.5.3 Connect Austria (Case C-462/99) [2003] ECR I-5197 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .16.2.1 Corporacíón Dermoestética (Cases C-500/06) [2008] ECR I-5785 . . . . . . . . . . . . . . . . . . . . . 11.5.2.4 Costa v Enel (Case 6/64) [1964] ECR 585 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .10.2, 11.2 De Danske Bilimportorer (C-383/01) [2003] ECR I-6065 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .11.4.3 De Peijper (Case 104/75) [1976] ECR 613 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11.5.2.1 Doux élevages SNC et al (Case C-677/11) 30 May 2013, not yet reported . . . . . . . . . . . . . . . . . .9.3.4 Eggers v Freie Hansestadt Bremen (Case 13/78) [1978] ECR 1935 . . . . . . . . . . . . . . . . . . . . . 11.5.2.1 Enel Produzione SpA v Autorità per l'energia elettrica e il gas (Case C-242/10) [2011] ECR I-13665 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.5 Essent Belgium NV v Vlaamse Reguleringsinstantie voor de Elektriciteits—en Gasmarkt (Joined Cases C-204/12 to C-208/12) 11 September 2014, not yet reported . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.5, 11.1, 11.4.4, 11.5.4, 11.5.4.1, 11.5.4.2 Essent Netwerk Noord v Aluminium Delfzijl (Case C-206/06) [2008] ECR I-5497 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9.3.4, 9.5.2, 11.2, 11.4.2, 11.4.3 Fazenda Pública v Fricarnes (Case C-28/96) [1997] ECR I-4939 . . . . . . . . . . . . . . . . . . . . . . . .11.4.3 Federutility and others v Autorità per l’energia elettrica e il gas (Case C-265/08) [2010] ECR I-3377 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.5, 9.3.3 France et al v Commission (Joined Cases T-139/09, T-243/09, and T-328/09) 27 September 2012, not yet reported . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .9.3.4 France v Commission (Case C-482/99) [2002] ECR I-4397 . . . . . . . . . . . . . . . . . . . . . . . . . . . .16.4.1 France v Ladbroke Racing and Commission (Case C-83/98 P) [2000] ECR I-3271 . . . . . . . . . . .9.3.4 Frans-Nederlandse Maatschappij voor Biologische Producten (Case 272/80) [1981] ECR 3277. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11.5.4.2 Fred Olsen (Case T-17/02) [2005] ECR II-2031 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.5 Frohnleiten v Austria (Case C-221/06) [2007] ECR I-9643 . . . . . . . . . . . . . . . . . . . . . . . . . . . . .11.4.3 Gambelli and Others (Case C-243/01) [2003] ECR I-13031 . . . . . . . . . . . . . . . . . . . . . . . . . . 11.5.2.4 Gas Natural Fenosa SDG v Commission (Case T-484/10 R) 17 February 2011, not yet reported . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.5 Geddo (Case 2/73) [1973] ECR 865 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .11.4.4 Germany v Commission (Case T-134/14) appeal brought on 28 February 2014, pending . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .9.3.4 Gilli and Andres (Case 788/79) [1980] ECR 2071 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11.5.2.1 Gouvernement de la Communauté française et Gouvernement wallon (Case C-212/06) [2008] ECR I-1683. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .14.4.1 Green Network SpA v Autorità per l’energia elettrica e il gas (Green Network) (Case C-66/13) 26 November 2014, not yet reported . . . . . . . . . . . . . . . . . . . . . . . . . . . 17.4.2.2 Groenveld (Case 15/79) [1979] ECR 3409 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .11.4.5

Table of Cases

xvii

Gysbrechts and Santurel (Case C-205/07) [2008] ECR I-9947 . . . . . . . . . . . . . . . . . . . . 11.4.5, 11.5.4 Haahr Petroleum (Case C-90/94) [1997] ECR I-4085 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .11.4.3 Hartlauer (Case C-169/07) [2009] ECR I-1721 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11.5.2.4 Humblot (Case 112/84) [1985] ECR 1367 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .11.4.3 Iannelli & Volpi v Meroni (Case 74/76) [1977] ECR 557 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11.3 Iride SpA and Iride Energia SpA v Commission (Case T-25/07) [2009] ECR II-245 . . . . . . . . . .9.3.2 Italy v Commission (Case 173/73) [1974] ECR 709. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .9.3.4 Jägerskiöld v Gustafsson (Case C-97/98) [1999] ECR I-7319 . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11.2 Keck (Case C-267/91) [1993] ECR I-6097 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .11.4.4 Larsen (Case 142/77) [1978] ECR 1543 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .11.4.3 LIBRO (Case C-531/07) [2009] ECR I-3717 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .11.4.4 Metock e.a. (Case C-127/08) [2008] ECR I-6241 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .14.4.1 Monsees (Case C-350/97) [1999] ECR I-2921 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .11.4.5 Oebel (Case 155/80) [1981] ECR 1993 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .11.4.5 Outokumpu Oy (Case C 213/95) [1998] ECR I-1777 . . . . . . . . . . . . . . . . . . . . .9.5.2, 11.4.3, 11.5.4.2 Pearle (Case C-345/02) [2004] ECR I-7139 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .9.3.4 Poste Italiane (Case T-525/08) 3 September 2013, not yet reported . . . . . . . . . . . . . . . . . . . . . . .9.3.1 PreussenElektra AG v Schhleswag AG (Case C-379/98) [2001] ECR I-2099 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9.3.4, 11.4.4, 11.5.4.2, 12.4.2, 13.4.2 Procureur du Roi v Dassonville (Case 8/74) [1974] ECR 837 . . . . . . . . . . . . . . . . . . . . . . . . . . .11.4.4 R v Secretary of State for the Home Department, Ex p Evans Medical (Case C-324/93) [1995] ECR I-563 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11.5.4.1 Radlberger v Land Baden-Württemberg (Case C-309/02) [2004] ECR I-11763 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11.4.4, 11.5.2.3 Ratti (Case 148/78) [1979] ECR 1629. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11.5.2.1 Rewe-Zentral (Cassis de Dijon) (Case 120/78) [1979] ECR 649 . . . . . . . . . . 11.5.2.1, 11.5.4, 11.5.4.2 Rewe-Zentralfinanz v Landwirtschaftskammer (Case 4/75) [1975] ECR 843 . . . . . . . . . . . . . . .11.5.3 Schutzverband gegen unlauteren Wettbewerb v TK-Heimdienst Sass (Case C-254/98) [2000] ECR I-151 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11.5.4.1 Simmenthal v Italian Minister of Finance (Case 35/76) [1976] ECR 1871 . . . . . . . . . . . . . . . 11.5.2.1 Sloman Neptun (Cases C-72 and 73/91) [1993] ECR I-887 . . . . . . . . . . . . . . . . . . . . . . . . . . . .16.4.1 Social Fonds voor de Diamantarbeiders v Brachfeld (Cases 2 and 3/69) [1969] ECR 211. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .11.4.2 Steinike & Wenlig v Germany (Case 76/78) [1977] ECR 595 . . . . . . . . . . . . . . . . . . . . . . . . . . . .9.3.4 Stoss (Case C-316/07) [2010] ECR I-8069 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11.5.2.4 Tasca (Case 65/75) [1976] ECR 291 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .11.4.4 Tedeschi v Denkavit (Case 5/77) [1977] ECR 1555 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11.5.2.1 Torfaen v B&Q (Case C-145/88) [1989] ECR 3831 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11.5.2.4 Van Tiggele (Case 82/77) [1978] ECR 25. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .11.4.4 Vereniging voor Energie, Milieu en Water and Others v Directeur van de Dienst uitvoering en toezicht energie (Case C-17/03) [2005] ECR I-04983 . . . . . . . . . . . . .1.5.4 OTHER TRIBUNALS (chronological order) Tribunal Supremo, Decision of 10 October 2007 in Appeal No 14/2006 . . . . . . . . . . . . . . . . . Tribunal Supremo, Decision of 28 January 2009 in Appeal No 37/2007. . . . . . . . . . . . . . . . . . Tribunal Supremo, Decision of 28 January 2009 in Appeal No 42/2007. . . . . . . . . . . . . . . . . . Tribunal Supremo, Decision of 7 March 2011 in Appeal No 92/2009 . . . . . . . . . . . . . . . . . . . Tribunal Supremo, Decision of 8 March 2011 in Appeal No 90/2009 . . . . . . . . . . . . . . . . . . . Tribunal Supremo, Decision of 18 March 2011 in Appeal No 623/2009 . . . . . . . . . . . . . . . . . Tribunal Supremo, Decision of 22 March 2011 in Appeal No 87/2009 . . . . . . . . . . . . . . . . . . Tribunal Supremo, Decision of 18 July 2013 in Appeal No 87/2011. . . . . . . . . . . . . . . . . . . . . Tribunal Supremo, Decision of 8 November 2013 in Appeal No 44/2012 . . . . . . . . . . . . . . . .

21.3.5 21.3.5 21.3.5 21.3.5 21.3.5 21.3.5 21.3.5 21.3.5 21.3.5

Table of Legislation (chronological order) 1. INTERNATIONAL TREATIES AND AGREEMENTS EC Switzerland Free Trade Agreement [1972] OJ L 300/189 . . . . . . . . . . . 17.4.2.2 Agreement on the European Economic Area [1994] OJ L 1/3 . . . . . . . . . . . . 11.4.4 Agreement between Denmark, Iceland, and Norway on cooperation in competition cases, 16 March 2001 . . . . . . . . . . 19.3.2.5 Agreement between the government of the Kingdom of Norway and the government of the Kingdom of Sweden on a common market for electricity certificates, 29 June 2011 (Swedish-Norwegian Treaty) . . . . 19.3.2.3 Treaty on the Functioning of the European Union [2012] OJ C 326/47 . . . . . 1.4.2, 1.5, 3.3.2, 9.3.3, 10.5, 11.1, 13.4.2, 14.4.1

2. EU LEGISLATION (1) Regulations Council Regulation (EC) No 659/1999 of 22 March 1999 laying down detailed rules for the application of Article 108 of the Treaty on the Functioning of the European Union of the EC Treaty [1999] OJ L 83/1 . . . . . . . . . . . . . . . . . . 9.5 Council Regulation (EC) No 1/2003 of 16 December 2002 on the implementation of the rules on competition laid down in Articles 81 and 82 of the Treaty [2003] OJ L 1/1 . . . . . . . . . . . . . . . . . . 10.2 Regulation (EC) No 1228/2003 of the European Parliament and of the Council of 26 June 2003 on conditions for access to the network for cross-border exchanges in electricity [2003] OJ L 176/1 . . . . . . 1.1 Regulation (EC) No 714/2009 of the European Parliament and of the Council of 13 July 2009 on conditions for access to the network for cross-border exchanges in electricity and repealing Regulation (EC) No 1228/2003 [2009] OJ L 211/

15 (2009 Cross-border Regulation) . . . 1.1, 2.1, 5.3.1, 5.5.2, 9.2, 9.4, 16.2.2, 19.4.1 Regulation (EU) No 1227/2011 of the European Parliament and of the Council of 25 October 2011 on wholesale energy market integrity and transparency [2011] OJ L 326/1 (REMIT) . . . . . . . . . 10.1, 10.5.1.5, 22.3.4.1 Commission Regulation (EU) No 360/2012 of 25 April 2012 on the application of Articles 107 and 108 of the Treaty on the Functioning of the European Union to de minimis aid granted to undertakings providing services of general economic interest [2012] OJ L 114/8 . . . . . . . . . . 9.3 Regulation (EU) No 347/2013 of the European Parliament and of the Council of 17 April 2013 on guidelines for trans-European energy infrastructure and repealing Decision No 1364/ 2006/EC and amending Regulations (EC) No 713/2009, (EC) No 714/2009 and (EC) No 715/2009 [2013] OJ L 115/39 (Regulation on trans-European energy infrastructure) . . . . . . . . . . . . . . . 1.5.1, 12.4 (Draft) Regulation establishing a Guideline on Capacity Allocation and Congestion Management, provisional final version, 5 December 2014 . . . . . . . . . . . . . . . . 5.3.1 Commission Regulation (EU) 651/2014 of 17 June 2014 declaring certain categories of aid compatible with the internal market in application of Articles 107 and 108 of the Treaty [2014] OJ L 187/1 . . . . . . . . 9.4 (2) Directives Council Directive 68/414/EEC of 20 December 1968 imposing an obligation on Member States of the EEC to maintain minimum stocks of crude oil and/or petroleum products [1968] OJ L 308/14 . . . . 11.5.2.1, 11.5.4.1 Council Directive 72/425/EEC of 19 December 1972 amending the Council Directive of 20 December 1968 imposing an obligation on Member States of the EEC

xx

Table of Legislation

to maintain minimum stocks of crude oil and/or petroleum products [1972] OJ L 291/154 . . . . . . . . . . . . . . . . . 11.5.4.1 Council Directive 73/238/EEC of 24 July 1973 on measures to mitigate the effects of difficulties in the supply of crude oil and petroleum products [1973] OJ L 228/1. . . . . . . . . . . . . . . . . . . 11.5.2.1 Directive 96/92/EC of the European Parliament and of the Council of 19 December 1996 concerning common rules for the internal market in electricity [1997] OJ L 27/20 (1996 Electricity Directive) . . . . . 1.1, 1.3.1, 15.2.3, 16.2.1, 17.2.2 Directive 98/30/EC of the European Parliament and of the Council of 22 June 1998 concerning common rules for the internal market in natural gas [1998] OJ L 204/1 (1998 Gas Directive) . . . . . . . . . . . . 15.2.3, 16.2.1 Directive 2000/60/EC of the European Parliament and of the Council of 23 October 2000 establishing a framework for the Community action in the field of water policy [2000] OJ L 327/1 (EU Water Framework Directive) . . . . . . 19.2.2 Directive 2001/77/EC of the European Parliament and of the Council of 27 September 2001 on the promotion of electricity from renewable energy sources in the internal electricity market [2001] OJ L 283/33 (2001 RES Directive). . . . . . . . . . 1.4.4, 11.4.4, 11.5.4.2 Directive 2001/80/EC of the European Parliament and of the Council of 23 October 2001 on the limitation of emissions of certain pollutants into the air from large combustion plants [2001] OJ L 309/1 . . . . . . . . . . . . . . . 13.2.3 Directive 2003/54/EC of the European Parliament and of the Council of 26 June 2003 concerning common rules for the internal market in electricity and repealing Directive 96/92/EC [2003] OJ L 176/37 (2003 Electricity Directive) . . . . . . .1.1, 1.3.1, 9.2, 11.4.4, 11.5.4.2, 17.2.2, 21.3.3, 21.3.5 Directive 2004/17/EC of the European Parliament and of the Council of 31 March 2004 coordinating the procurement procedures of entities

operating in the water, energy, transport, and postal services sector [2004] OJ L 134/1 . . . . . . . . . . . . . . . 19.2.2 Directive 2005/89/EC of the European Parliament and of the Council of 18 January 2006 concerning measures to safeguard security of electricity supply and infrastructure investment [2006] OJ L 33/22 (Security of Supply Directive). . . . . . . . . . . 2.4, 3.4.3, 5.5.2, 6.1, 9.2, 13.4.2, 14.4.1, 16.4.3 Directive 2009/28/EC of the European Parliament and of the Council of 23 April 2009 on the promotion of the use of energy from renewable sources and amending and subsequently repealing Directives 2001/77/EC and 2003/30/EC [2009] OJ L 140/16 (2009 RES Directive) . . . . . . . 1.4, 1.4.4, 11.4.4, 16.2.3, 20.2.3 Directive 2009/72/EC of the European Parliament and of the Council of 13 July 2009 concerning common rules for the internal market in electricity and repealing Directive 2003/54/EC [2009] OJ L 211/55 (2009 Electricity Directive). . . . . . . . . . . 1.1, 1.1.2, 1.4.1, 1.5, 1.5.3, 2.1, 3.3.2, 5.3.1, 6, 9.3.3, 13.3.1, 13.3.2.1, 13.4.1, 14.4.1, 16.2.2, 16.4.3, 17.2.2, 17.4.1, 18.2.2, 19.2.2, 19.4.1, 20.2.2, 20.2.3, 21.2.5, 21.3.6 Directive 2009/73/EC of the European Parliament and of the Council of 13 July 2009 concerning common rules for the internal market in natural gas and repealing Directive 2003/55/EC [2009] L 211/94 (2009 Gas Directive). . . . . . . . . . . . . . 2.1, 9.3.3, 19.4.1 Council Directive 2009/119/EC of 14 September 2009 imposing an obligation on Member States to maintain minimum stocks of crude oil and/or petroleum products [2009] OJ L 265/9 . . . . . 11.5.2.1 Directive 2010/75/EU of the European Parliament and of the Council of 24 November 2010 on industrial emissions (integrated pollution prevention and control) [2010] OJ L 334/17 . . . . . . . 20.2.3 Directive 2012/27/EU of the European Parliament and of the Council of 25 October 2012 on energy efficiency, amending Directives 2009/125/EC and 2010/30/EU and repealing Directives

Table of Legislation 2004/8/EC and 2006/32/E Commission Directive 2012/27/EU on energy efficiency [2012] OJ L 315/1 . . . . . . . . 12.4 (3) Decisions

Council Decisions (chronological order) 77/186/EEC: Council Decision of 14 February 1977 on the exporting of crude oil and petroleum products from one Member State to another in the event of supply difficulties [1977] OJ L 61/23 . . . . 11.5.2.1 77/706/EEC: Council Decision of 7 November 1977 on the setting of a Community target for a reduction in the consumption of primary sources of energy in the event of difficulties in the supply of crude oil and petroleum products [1977] OJ L 292/9 . . . . . . . . . . . . . . . . . . . . . . . 11.5.2.1

Commission Decisions (alphabetical order) 2012/21/EU: Commission Decision of 20 December 2011 on the application of Article 106(2) of the Treaty on the Functioning of the European Union to State aid in the form of public service compensation granted to certain undertakings entrusted with the operation of services of general economic interest (notified under document C(2011) 9380) [2012] OJ L 7/3 . . . . . . . . . . . . . . 9.3 AB Klaipėdos nafta (Lithuania LNG Terminal) (Case SA.36740) 20 November 2013, not yet reported . . . . . . . . . . . . . 9.3.3, 9.3.4, 9.5 Aid for Capacity Payments for Oil-Shale Fuelled Electricity Production (ET) (Case SA.30531) [2011] OJ C/235 . . . . . . . . 1.4.3 Austria—Green Electricity Act (Case C 24/09, ex NN 446/08) [2011] OJ L 235 . . . . . . . 9.3.4 CHP certificates (Case N 608/2004) [2005] OJ C 240 . . . . . . . . . . . . . . . . 12.4.2 Costs of transition to competition (Spanish stranded costs) (Case SA NN 49/99) [2001] OJ C 268 . . . . . . . . . . . . . . . . . 21.4 Distrigaz (Case COMP/37.966) [2008] OJ C 9 . . . . . . . . . . . . . . . . . . . . . . . . . 10.2 Long term electricity contracts in France (EDF) (Case COMP/39.386) [2010] OJ C 133/5 . . . . . . . . . . . . . . . . . . . . . 10.2

xxi

Long term electricity contracts in Belgium (Electrabel) (Case COMP/39.387) no decision, proceedings closed on 3 February 2011 . . . . . . . . . . . . . . . . . 10.2 ENI (Case COMP/39.315) [2010] C 352/8 . . . . . . . . . . . . . . . . . . . . . . . . 10.2 E.ON/MOL (Case COMP/M.3696) [2006] OJ L 253/20 . . . . . . . . . . . . . . . 10.4 Exeltium (MEMO/08/533 of 31 July 2008) . . . . . . . . . . . . . . . . . . . . 10.2 French highways (Case N 540/2000) [2000] C 354 . . . . . . . . . . . . . . . . . . . 12.4.2 Gazprom/ENI (Case COMP/38.308) settled on 6 October 2003 (Press release IP/03/1345) . . . . . . . . . . . . . . . 10.2 Gazprom/E.ON Ruhrgas (Case COMP/38.307) settled on 10 June 2005 (Press release IP/05/710) . . . . . . . . . . . . . . . . . . . . . . 10.2 Gazprom/OMV (Case COMP/38.085) settled on 17 February 2005 (Press release IP/05/195) . . . . . . . . . . . . . . . . 10.2 GDF (COMP/39.316) [2010] C 57/13 . . . . 10.2 GDF/ENEL (Case COMP/38.662) 26 October 2004 . . . . . . . . . . . . . . . . . 10.2 German Electricity Wholesale Market and German Electricity Balancing Market (Cases COMP/39.388 and COMP/ 39.389) [2009] OJ C 36/8 (E.ON cases) . . . . . . . . . . .10.2, 10.3, 10.4, 10.5.1, 10.5.1.3 Germany—Support for renewable electricity and reduced EEG-surcharge for energy-intensive users (Case SA.33995 (2013/C) (ex 2013/NN)) C(2014) 8786 final (EEG-surcharge) . . . . 9.3.4, 9.3.5, 9.5, 9.5.2, 15.2.1 Green certificates for promoting electricity from renewable sources (Photovoltaics Romania) (Case SA.33134) [2011] OJ C 244 . . . . . . . . . . . . . . . . 12.4.2, 14.4.3 Ireland, Public Service Obligation—Electricity Supply Board (Case N 143/2004) [2005] OJ C/242 . . . . . . . . . . . . . . . . . 1.3.2 Irish CADA (Case N 475/2003) [2003] OJ C 34/7 . . . . . . . . . . . . . . . . . .1.3.2, 20.4 Netherlands – NOx Trading Scheme (N 35/ 2003) [2003] OJ C 227/8. . . . . . . . . . 12.4.2 RWE (Case COMP/39.402) [2008] OJ C 133/10 . . . . . . . . . . . . . . . . . . . . 10.2 Sonatrach (Case COMP/37.811) settled on 11 July 2007 (Press release IP/07/1074) . . . . . . . . . . . . . . . . . . . . . 10.2

xxii

Table of Legislation

State aid implemented by the Czech Republic for several regional bus service operators in the Ústí Region (Czech buses) (Case SA.20350) [2014] OJ L 329/35 . . . . . . 9.3.3 Swedish Interconnectors (COMP/39.351) [2010] C 142/28 . . . . . . . . . . . . . . . . . 10.2 Tender for Aid for New Electricity Generation Capacity (LV) (Case 675/2009) [2010] OJ C/213 . . . . . . . . . . . . . . . . . . 1.4.2, 1.5.1 United Kingdom Electricity Market Reform—Capacity Market (Case SA.35980) [2014] OJ C/348 . . . . . . . . . . . 1.2.3.4, 1.5.1, 1.5.2, 1.5.4, 9.3.4, 9.4.1, 9.5, 9.5.1, 9.5.2, 10.1, 10.3, 20.3.2.2, 20.4, 22.4.1, 22.4.2 United Kingdom Electricity Market Reform—Contract for Difference for Renewables (Case SA.36196 (2014/N)) C(2014) 5079 final [2014] OJ C 393 . . . . . . . . . . . . . . . . . . . . . . . 9.5.2 United Kingdom Investment Contract (early Contract for Difference) for the Hinkley Point C New Nuclear Power Station (Hinkley Point) (Case SA.34947 (2013/C) (ex 2013/N)) [2014] OJ C/69 . . . 9.3.3, 9.5 United Kingdom Feed-in Tariffs (Case N94/ 2010) 14 April 2010 . . . . . . . . . . . . . . 9.3.4 United Kingdom Support for five Offshore Wind Farms: Walney, Dudgeon, Hornsea, Burbo Bank and Beatrice (Cases SA.38758 (2014/N), SA.38759 (2014/N), SA.38761 (2014/N), SA.38763 (2014/N) and SA.38812 (2014/N)) [2014] OJ C 393 . . . . . . . . . . . . . . . . . 9.5.2

Other EU Decisions 1364/2006/EC: Decision of the European Parliament and of the Council of 6 September 2006 laying down guidelines for trans-European energy networks and repealing Decision 96/391/EC and Decision 1229/2003/EC [2006] OJ L 262/1 . . . . . . . . . . . . . . . . . . . 12.2.1, 21.2.1 3. NATIONAL LEGISLATION (Divided into primary legislation, secondary legislation, and acts of the National Regulatory Authority, within one subset acts are arranged by decreasing importance as a source of law, and for acts of equal ranking—chronologically)

(1) EU COUNTRIES

Austria Bundesverfassungsgesetz, mit dem die Eigentumsverhältnisse an den Unternehmen der österreichischen Elektrizitätswirtschaft geregelt warden 1998 (Federal Constitutional Law on the ownership of corporations of the Austrian electricity industry) BGBl. I Nr. 143/1998. . . . . . . . . . . . . . . . . . 12.2.1 Ökostromgesetz (ÖSG) 2002, BGBl. I Nr. 149/2002 . . . . . . . . . . . . . . . . . . . 12.2.1 Elektrizitätswirtschafts- und – organisationsgesetz (ElWOG) 2010, BGBl. I Nr. 110/2010 . . . . . . 12.2.2, 12.3.1 Ökostromgesetz (ÖSG) 2012 – Bundesgesetz über die Förderung der Elektrizitätserzeugung aus erneuerbaren Energieträgern, BGBl. I Nr. 75/2011, as amended BGBl. I Nr. 11/2012 . . . 12.2.2.1 Energielenkungsgesetz (EnLG) 2012, BGBl. I Nr. 41/2013 . . . . . . . . . . . . . . . . 12.2.2, 12.2.2.1 Bundesgesetz über die Regulierungsbehörde in der Elektrizitäts- und Erdgaswirtschaft, Energie-Control-Gesetz (E-ControlG) 2013, BGBl. I Nr. 174/2013. . . . . . . . . . . 12.2.2, 12.2.2.3

Belgium Primary legislation Federal Act regarding the organization of the electricity market of 29 April 1999, Belgian Official Gazette, 11 May 1999 (Electricity Act) . . . . . . . . . . . . . 13.2.2, 13.3.1, 13.3.2.2 Nuclear Phase-out Act of 31 January 2003, Belgian Official Gazette, 28 February 2003. . . . . . . . . . . . . . . . . . . . . . . . . . 13.2.1 Law of 18 December 2013 amending the Nuclear Phase-out Act of 31 January 2003 and the Act of 11 April 2003 regarding the provisions for the dismantling of the nuclear power plants, Belgian Official Gazette, 24 December 2013 . . . . . . . 13.2.3 Law of 26 March 2014 amending the Electricity Act, Belgian Official Gazette, 1 April 2014 (Strategic Reserve Law). . . . . . . . . 13.3.2.2 Secondary legislation Royal Decree of 8 December 2013 concerning the conditions of the tendering procedure under Art 5 of the

Table of Legislation Electricity Act, Belgian Official Gazette, 23 December 2013. . . . . . . . . . . . . 13.3.2.1 Ministerial Decision of 18 November 2013 concerning the use of the tendering procedure pursuant to Art 5(2) of the Electricity Act, Belgian Official Gazette, 2 December 2013 . . . . . . . . . . . . . 13.3.2.1 Ministerial Decision of 3 April 2014. . . . . . . . . . . . . . . . . . . . . . . . 13.3.2.2 Ministerial Decision of 16 July 2014. . . . . . . . . . . . . . . . . . . . . . . . 13.3.2.2

xxiii

Germany Energiewirtschaftsgesetz (EnWG) of 7 July 2005, BGBl. I S. 1970, 3621, last amended by BGBl. I S. 1066 . . . .15.2.3, 15.3.1, 15.3.2.2, 15.3.2.3, 15.3.2.5, 15.3.2.6 Reservekraftwerksverordnung (ResKV) of 27 July 2013, BGBl. I S. 1947 . . . 15.3.1, 15.3.2.2, 15.3.2.3, 15.4 Erneuerbare-Energien-Gesetz (EEG) of 21 July 2014, BGBl. I S. 1066, last amended by BGBl. I S. 2406 . . . . . . 9.3.4, 15.2.1, 15.2.3

Acts of the National Regulatory Authority CREG, Decision (B)130626-CDC-1248 of 26 June 2013 regarding reserve capacity for 2014 . . . . . . . . . . . . . . . . . . . . . . . 13.3.1

France Primary legislation Law No 2000–108 of 10 February 2000 on the modernization and development of the public electricity service, consolidated version of 1 January 2012 . . . . . . . . . . . 6.1 French Energy Code, Law of 13 July 2005, consolidated version of 1 January 2015. . . . . . . . . . 14.2.1.1, 14.3.1.2, 14.3.1.3 Law 2010–1488 of 7 December 2010 on the New Organisation of the Electricity Market (Loi no 2010–1488 du 7 décembre 2010 portant nouvelle organisation du marché de l’électricité) (NOME law) . . . . . . . 1.2.3.3, 14.3, 14.3.1.1 Law No 2013–312 of 15 April 2013 (Loi n 2013–312 du 15 avril 2013 visant à préparer la transition vers un système énergétique sobre et portant diverses dispositions sur la tarification de l’eau et sur les éoliennes) French Official Journal, 16 April 2013 . . . . . . . . . . . . . . .1.5.2, 14.3 Secondary legislation Ministerial Decrees of 10 January 2010, French Official Journal . . . . . . . . . . . . . . . 14.2.1.1 Decree 2011–466 of 28 April 2011 (Décret no 2011–466 du 28 avril 2011 fixant les modalités d’accès régulé à l’électricité nucléaire historique) French Official Journal. . . . . . . . . . . . . . . . . . . . . . 14.3.1.3 Decree 2012–1405 of 14 December 2012, French Official Journal (Capacity Decree) . . . 14.3, 14.3.1.3, 14.3.2, 14.3.2.1, 14.3.2.2, 14.3.2.3, 14.4.1, 14.4.3

Greece Primary legislation Law 2773/1999, Government’s Gazette, Issue No 286 (GG A 286) 22 December 1999. . . . . . . . . . . . . . . . . . . . . . . . . . 16.2.1 Law 3175/2003, Government’s Gazette, Issue No 207 (GG A 207) 29 August 2003. . . . . . . . . . . . . . . . . . . . . . . . . . 16.3.1 Law 4001/2011, Government’s Gazette, Issue No 179 (GG A 179) 22 August 2011 . . . 16.2.1 Secondary legislation Ministerial Decision ΥΑ Δ13/Φ.7.1/13624, Government’s Gazette, Issue No 1058 (GG B 1058) 4 August 2014 . . . . . . . 16.4.3 Acts of the National Regulatory Authority RAE, Grid and Market Operation Code, Government’s Gazette, Issue No 665 (GG B 655) 9 May 2005 . . . . . . . . . 1.2.3.3, 16.2.2, 16.3.1, 16.3.2, 16.4.2, 16.4.3 RAE, Decision 56/2012, Government’s Gazette, Issue No 104 (GG B 104/2012) (the Greek Exchange Code) . . . . . . . 16.3.1 RAE, Decision 57/2012, Government’s Gazette, Issue No 103 (GG B 103/2012) (the Greek Transmission System Operation Code) . . . . . . . . . . . . . . . . 16.3.1 RAE, Decision 338/2013, Government’s Gazette, Issue No 1795 (GG B 1795) 25 July 2013. . . . . . . . . . . . . . . . . . . 16.2.1, 16.2.2, 16.3.2, 16.4.1 RAE, Decision 339/2013, Government’s Gazette, Issue No 1795 (GG B 1795) 25 July 2013 . . . . . . . . . . . . . 16.2.1, 16.2.2, 16.3.2, 16.4.1, 16.4.3

xxiv

Table of Legislation

Ireland Eirgrid, Grid code version 5.0, October 2013. . . . . . . . . . . . . . . . . . . . . . . . . . . 3.2.2

Italy Law No 239/2004 of 23 August 2004 (Marzano Law). . . . . . . . . . . . . . . . . . . . 17.2.2, 17.3.1 Law No 147/2013 of 27 December 2013. . . . . . . . . . . . . . . . . . . 17.3.1, 17.3.2.1 Legislative Decree No 79/1999 of 16 March 1999 (Bersani Decree) . . . . . . . . . 17.2.2, 17.4.2.2 Legislative Decree No 379/2003 of 19 December 2003 . . . . . . . 17.2.2, 17.3.2.3 Ministerial Decree of 11 November 1999. . . . . . . . . . . . . . . . . . . . . . . . 17.4.2.2 Ministerial Decree of 8 March 2013 . . . . . 17.4.1 Acts of the National Regulatory Authority AEEG, Resolution ARG/elt 98/11 of 21 July 2011, Criteri e condizioni per la disciplina del sistema di remunerazione della disponibilità di capacità produttiva di energia elettrica, ai sensi dell’articolo 2 del decreto legislativo 19 dicembre 2003, n. 379 (capacity market) . . . . . . . . . . . 7.1, 17.2.2 AEEG, Decision 48/2004 of 27 March 2004. . . . . . . . . . . . . . . . . . . . . . . . . . 17.3.1 AEEG, Decision 98/2011 of 21 July 2011 . . . . .17.3.2, 17.3.2.1, 17.3.2.2 AEEG, Decision 482/2012 of 15 November 2012. . . . . . . . . . . . . . . . . . . . . . . . 17.3.2.1 AEEG, Decision 375/2013 of 5 September 2013. . . . . . . . . . . . . . . . . 17.3.2.1, 17.3.2.2

Netherlands Elektriciteitswet, 2 July 1998, Stb. 1998, 427, as amended Stb. 2013, 325 (Electricity Act). . . . . . . . . . . . . . . . . 18.2.2

Norway Lov om erverv av vannfall mv. (industrikonsesjonsloven), LOV-191712-14-16 (Industrial Licence Act). . . 19.2.2 Lov om vasdragsreguleringer (vassdragsreguleringsloven), LOV-1917-12-14-17 (Water Regulation Act). . . . . . . . . . . 19.2.2

Lov om offentlige anskaffelser (anskaffelsesloven), LOV-1999-07-16-69 (Procurement Act) . . . . . . . . . . . . . . 19.2.2 Lov om vassdrag og grunnvann (vannressursloven), LOV-2000-11-24-82 (Water Resources Act). . . . . . . . . . . . 19.2.2 Lov om produksjon, omforming, overføring, omsetning, fordeling og bruk av energi m.m. (energiloven), Lov-2013-06-14-53 (Energy Act) . . . . . . . . . . . . 19.2.2, 19.3.2.4 Lov om elsertifikater, LOV-2013-12-13-126 (Electricity Certificate Act) . . . . . . 19.3.2.3, 19.3.2.5 Forskrift om produksjon, omforming, overføring, omsetning, fordeling og bruk av energi m.m. (energilovforskriften), FOR-1990-12-07-959 (Energy Regulation) . . . . . . . . . . . . . 19.2.2 Forskrift om måling, avregning og samordnet opptreden ved kraftomsetning og fakturering av nettjenester, FOR-1999-0311-301 (Settlement Regulation) . . . . 19.2.2 Forskrift om økonomisk og teknisk rapportering, inntektsramme for nettvirksomheten og tariffer, FOR1999-03-11-302 (Regulation for Reporting and Revenue Cap) . . . . . . 19.2.2 Forskrift om elsertifikater, FOR-201112-16-1398 (Electricity Certificate Regulation) . . . . . . . . . . . . . . . . . . 19.3.2.3

Poland Primary legislation Energy Law Act 1997, Journal of Laws of 2012, item 1059, as amended (Ustawa z dn. 10 kwietnia 1997 r., Prawo energetyczne, Dz.U. z 2012 r., poz. 1059 z późn. zm.) . . . . . . . . . . . . . . . 20.2.2 Act of 26 July 2013 amending the Energy Law Act and related acts, Journal of Laws of 27 August 2013, item 984 . . . . . . . . . . . . . . . . . . . . . . 20.2.3 Act of 11 July 2014 amending the Environmental Law and Other Acts, Journal of Laws 2014, item 1101 . . . 20.2.3 Secondary legislation Regulation of the Minister of the Environment from 4 November 2014 regarding emission standards for certain types of installations, combustion plants, incineration and

Table of Legislation co-incineration plants, Journal of Laws 2014, item 1546 . . . . . . . . . . . . 20.2.3 PSE, Grid Code (Instruction of Operation and Exploitation of the Transmission Network (Instrukcja Ruchu i Eksploatacji Sieci Przesyłowej—IRiESP), version of 1 August 2014 . . . . . . . . . . . 20.2.3, 20.3.1.2 PSE, Karta Aktualizacji nr CB/9/2013, IRiESP—Balancing . . . . . . . . . . . . 20.3.1.2 Acts of the National Regulatory Authority PSE, Grid Code (Instruction of Operation and Exploitation of the Transmission Network (Instrukcja Ruchu i Eksploatacji Sieci Przesyłowej—IRiESP), version of 1 August 2014 . . . . . . . . . . . 20.2.3, 20.3.1.2 Decision of the President of the URE of 10 December 2013 (DRR-4320-2(27)/ 2010/2013/JRz) approving the amendment to the Grid Code (PSE, Karta Aktualizacji nr CB/9/2013, IRiESP—Balancing) . . . . . . . . . . . 20.3.1.2

Spain Primary legislation Electricity Sector Act No 54 of 27 November 1997, Spanish Official Bulletin No 285 of 28 November 1997 (ESA 1997) . . . . . . . . 21.2.2, 21.3.2, 21.3.5 Act No 15 of 3 July 2007 on the Protection of Competition, Spanish Official Bulletin No 159 of 4 July 2007 . . . . . . . . . . . . 21.3.5 Electricity Sector Act No 24 of 26 December 2013, Spanish Official Bulletin No 310 of 27 December 2013 (ESA 2013) . . . . . . . . 21.2.2, 21.2.3, 21.2.5, 21.3.1, 21.3.6

xxv

Royal Decree No 1955/2000 of 1 December 2000, Spanish Official Bulletin No 310 of 27 December 2000 . . . . . . . . . . . . 21.2.5 Royal Decree No 1454/2005 of 2 December 2005, Spanish Official Bulletin No 306 of 23 December 2005 . . . . . . . . . . . . 21.3.5 Royal Decree No 1634/2006 of 29 December 2006, Spanish Official Bulletin No 310 of 30 December 2006 . . . . . . . . . . . . 21.3.5 Ministerial Order of 17 December 1998 developing some aspects of Royal Decree No 2019/1997, Spanish Official Bulletin No 310 of 28 December 1998 (MO 1998) . . . . . . . . . . 21.3.2, 21.3.5 Ministerial Order No ITC/2794 of 27 September 2007, Spanish Official Bulletin No 234 of 29 September 2007 (MO 2794/2007) . . . . . . . . . . 21.3.3, 21.3.4, 21.3.5 Ministerial Order No ITC/3860 of 28 December 2007, Spanish Official Bulletin No 312 of 29 December 2007 . . . . . . 21.3.3 Ministerial Order No ITC/3801 of 26 December 2008, Spanish Official Bulletin No 15 of 31 December 2008 . . . . . . . 21.3.3 Ministerial Order No ITC/3127/2011 of 17 November 2011, Spanish Official Bulletin No 278 of 18 November 2011 (MO 3127/ 2011) . . . . . . . . . . . . . . . . . . . . . . . . . 21.3.3 Ministerial Order No IET/2013/2013 of 31 October 2013, Spanish Official Bulletin No 262 of 1 November 2013 . . . . . . . 1.2.2 Decision of the General Secretary for Energy of 11 February 2005 approving REE’s Operating Procedure 12.1 (Applications for the connection to new installations to the transmission network), Spanish Official Bulletin No 51 of 1 March 2005. . . . . . . . . . . 21.2.5

Secondary legislation Royal Decree-Law No 13/2012 of 30 March 2012, Spanish Official Bulletin No 78 of 31 March 2012 (RDL 13/2012) . . . . . . . . . . . . . . . . . 21.3.4 Royal Decree-Law No 9/2013 of 12 July 2012, Spanish Official Bulletin No 167 of 13 July 2013 (RDL 9/2013) . . . . . . . . 21.3.4 Royal Decree No 2019/1997 of 26 December 1997, Spanish Official Bulletin No 310 of 27 December 1997 . . . . . . . . . . . . 21.3.2

UK Primary legislation Electricity Act 1989 (c. 29) . . . . . 22.2.2, 22.3.4.1 Utilities Act 2000 (c. 27) . . . . . . . . . . . . . . 22.2.2 Energy Act 2004 (c. 29) . . . . . . . . . . . . . . 22.2.2 Energy Act 2008 (c. 32) . . . . . . . . . . . . . . 22.2.2 Energy Act 2010 (c. 27) . . . . . . . . . . . . . . 22.2.2 Energy Act 2011 (c. 16) . . . . . . . . . . . . . . 22.2.2 Energy Act 2013 (c. 32) . . . . . . . . 22.2.2, 22.3.2

xxvi

Table of Legislation

Secondary legislation Electricity Capacity Regulations 2014, SI 2014/ 2043, 31 July 2014. . . . . . . . 22.3.2, 22.3.3.6 Electricity Capacity (Supplier Payment etc.) Regulations 2014, SI 2014/3354, 17 December 2014 . . . . . . . 22.3.2, 22.3.3.6 Capacity Market Rules 2014 (with amendments). . . . . . 10.5.1.4, 22.3.2, 22.3.3.2

cual se adopta la metodología para la remuneración del Cargo por Confiabilidad en el Mercado Mayorista de Energía, 3 October 2006 . . . . . . . . . . . . . 7.6.2, 7.6.4 Comisión de Regulación de Energía y Gas, Resolución No 148 de 2011, por la cual se define la metodología para determinar la energía firme de plantas eólicas, 21 October 2011 . . . . . . . . . . . . . . . . . 7.6.4

Acts of the National Regulatory Authority OFGEM, Decision of 19 December 2013, empowering the national grid to introduce new balancing services including payments to firms for creating reserves on the demand side . . . . . . . . . . . . . . . . . 22.2.3.3 OFGEM, Electricity Balancing Significant Code Review—Final Policy Decision, 15 May 2014 . . . . . . . . . . . . . . . . . 22.2.3.4 (2) NON-EU COUNTRIES

Colombia Comisión de Regulación de Energía y Gas, Resolución No 071/2006 por la

Peru Decree No 027/2002/PCM . . . . . . . . . . . . . . 7.4 Decree No 095/2003/EF . . . . . . . . . . . . . . . . 7.4

US ISO New England, Forward capacity market rules (FCM 101, 2014) . . . . . . . . . . . . . 7.3 PJM Capacity Market Operations, Manual 18 on PJM capacity market, Revision No 25, 30 October 2014 . . . . . . . . . . . . 7.3

Table of Relevant Non-Legislative Documents EU DOCUMENTS (documents issued by EU institutions and agencies, chronological order) European Commission, DG Competition report on the energy sector inquiry, SEC(2006) 1724, 10 January 2007. . . . . . 10.2, 10.5.1.3 Communication from the Commission on the application of the European Union State aid rules to compensation granted for the provision of services of general economic interest [2012] OJ C 8/4, 11 January 2012. . . . . . . . . . . . . . . . . . . . . . . . . 1.5, 9.3 Communication from the Commission, European Union framework for State aid in the form of public service compensation (2011) [2012] OJ C 8/15, 11 January 2012. . . . . . . . . . . . . . . . . . . . . . . 9.3, 9.3.3 Communication from the Commission, EU State aid modernisation (SAM), COM 2012 (209) final, 8 May 2012. . . . . . . . 9.2, 9.3.4 Communication from the Commission, Making the internal energy market work, COM (2012) 663 final, 15 November 2012 (2012 Internal Energy Market Communication). . . . . . . . . . . . .1.1, 19.4.3 European Commission, Consultation Paper on generation adequacy, capacity mechanisms and the internal market in electricity, 15 November 2012 (2012 Consultation Paper) . . . . . . . . 1.4.5, 7.1, 10.1, 12.3, 15.2.4, 19.4.2, 19.4.3 Opinion of the Agency for the Cooperation of Energy Regulators No 05/2013 of 15 February 2013 on capacity markets (ACER’s Opinion). . . . . . . . . . .2.1, 2.3, 2.4 European Council, Conclusions, EUCO 75/1/13 REV 1, 22 May 2013 . . . . . . . 2.1, 5.3.1, 9.1 ACER, Report on capacity remuneration mechanisms and the internal market for electricity, 30 July 2013 (ACER’s Report) . . . . . . . .1.2.3, 2.1, 3.4.1, 6.1, 11.1, 11.4.2, 11.6, 17.2.2, 17.4.2.1, 21.3.1

Communication from the Commission, Delivering the internal electricity market and making the most of public intervention, C(2013) 7243 final (November 2013 Communication). . . . . . . . 1.2.3.1, 1.4, 1.5, 2.1, 3.3.2, 5.5.4, 9.1, 9.2, 9.4, 11.6, 15.4, 17.4.2.1, 18.3.1 Commission Staff Working Document, Generation Adequacy in the internal electricity market—guidance on public interventions, SWD(2013) 438 final, 5 November 2013 (Generation Adequacy SWD) . . . . . 1.2.2, 1.4, 1.5, 1.5.1, 1.5.3, 1.5.4, 1.5.5, 1.6, 2.1, 4.5.2, 5.5.2, 5.5.4, 6.1, 6.4.1, 7.1, 9.2, 13.4.1, 15.4, 16.3.3, 17.4.2.1, 17.4.2.2 Communication from the Commission, Draft Commission Notice on the notion of State aid pursuant to Art 107(1) TFEU, issued for consultation on 17 January 2014. . . . . . . . . . . . . . . . . 9.3.2, 9.3.3, 9.3.4 Communication from the Commission, Guidelines on State aid for environmental protection and energy 2014–2020 [2014] OJ C 200/1, 28 June 2014 . . . . .1.4.5, 1.6, 2.1, 4.5.2, 5.1, 5.4, 5.4.2, 5.5.3, 6.1, 9.1, 9.3.3, 9.4, 9.4.1, 9.5.2, 12.4.2, 13.4.1, 16.4.1, 17.4.2.1, 17.4.2.2, 20.4 Communication from the Commission, Progress towards completing the Internal Energy Market, COM (2014) 634 final, 13 October 2014 (October 2014 Communication). . . . . . . . . . .9.2, 9.6, 11.6 European Commission, EU Energy Markets in 2014 (adapted version of the Commission Staff Working Documents SWD (2014) 310 final and SWD (2014) 311 final, accompanying the October 2014 Communication) 13 October 2014 . . . . . . . . . . . . . . . . 17.2.1

xxviii

Table of Relevant Non-Legislative Documents OTHER DOCUMENTS

1. EU Level (chronological order) ENTSO-E, Communication paper on capacity remuneration mechanisms, WG-RES, June 2012 . . . . . . . . . . . . . . . . . . . . . . . 6.1 THEMA consulting group, Capacity mechanisms in individual markets within the IEM, Report for the DirectorateGeneral Energy of the European Commission, June 2013 . . . . . . . . . 3.3.1.1, 21.3.2, 21.3.3, 21.4 ENTSO-E, Scenario Outlook and Adequacy Forecast 2014–2030, June 2014. . . . . 1.5.1, 3.3.1.1, 6.2.4, 9.2, 12.2.3, 16.2.3, 18.2.3 ENTSO-E, Ten-Year Network Development Plan (TYNDP) 2014 . . . . . . . . 5.5.1, 12.2.1 ENTSO-E, Target Methodology for Adequacy Assessment (Updated Version after Consultation) 14 October 2014 . . . . 16.3.3

2. Member State Level (countries are listed in alphabetical order, documents are listed in chronological order within one country) Austria Work Programme of the Austrian Federal Government 2013–2018, December 2003. . . . . . . . . . . . . . . . . . . . 12.4.1, 12.4.2 E-Control, Monitoring Report 2013 . . . . . 12.2.3 E-Control, Market Report 2014 . . . . . . . . 12.2.1 Belgium Federal Energy Administration (Algemene Directie Energie/Administration de l’Energie), Forecast Study for years 2008–2017, published 2009 . . . . . . . . . . . 13.2.2, 13.2.3 CREG, Capacity remuneration mechanism, Study (F)121011-CDC-1182, 11 October 2012 (CREG’s study on capacity mechanisms) . . . . . . .1.2.3.1, 13.2.1, 13.3.1, 13.3.2.2, 13.4.2, 17.3.1 Energy Policy Plan (Le système électrique belge à la croisée des chemins: une nouvelle politique énergétique pour réussir la Transition, Plan Wathelet) 27 June 2015 . . . . .13.3.2, 13.3.2.1, 13.3.2.2

France The Sido-Poignant Report, April 2010 . 14.2.1.2 Assemblée nationale, Impact Assessment of the NOME Law (Projet de loi portant nouvelle organisation du marché de l’électricité (NOME)—Etude d’impact) April 2010 . . . . . . . . . . . . . . . . . . . 14.3.1.1 Autorité de la concurrence, Opinion no 12-A-09 of 12 April 2012 . . . . . . . . . . 14.4.1, 14.4.2 RTE, French capacity market. Report accompanying the draft rules (Mécanisme de capacité: proposition de règles et dispositions complémentaires) 9 April 2014 . . . . . . . . . . . 3.3.1, 3.4.1, 7.1, 7.3.1.1, 7.6.2, 14.3.3 Germany BMWi, Energy Concept for an Environmentally Sound, Reliable and Affordable Energy Supply 2010–2011, 28 September 2010 (Energy Concept) . . . . . . . . . . . . . . . 15.2.1 BMWi, Report of the Power Plant Forum to the Federal Chancellor and the MinisterPresidents of the Länder, 28 May 2013. . . . . . . . 15.2.1, 15.2.2, 15.2.3, 15.2.4, 15.3.1, 15.3.4, 15.3.4.1, 15.3.4.2, 15.4 BNetzA, Bericht zum Zustand der leitungsgebundenen Energieversorgung im Winter 2012/13, 20 June 2013 . . . . . 15.2.1 EEX, Positionspapier der European Energy Exchange AG, Notwendigkeit und Design von Kapazitätsmechanismen, 5 August 2013 . . . . . . . . . . . . . . . . 15.3.4.3 BMWi, Ein Strommarkt für die Energiewende— Diskussionspapier des Bundesministeriums für Wirtschaft und Energie (Grünbuch) November 2014. . . . . . . . . . . 15.1, 15.3.4.1 Greece RAE, Announcement of 18 July 2013 regarding the adoption of a set of transitional regulatory measures . . . . . . . . . . . . . 16.2.2 ADMIE, Generation Adequacy Study for the period 2013–2020, October 2013. . . . . . . . . . . . . . . . . . . . 16.2.3, 16.3.3 RAE, Annual Report to the European Commission, December 2013. . . . . . . . . . . . . . . . . . . . 16.2.3, 16.3.2 RAE, Proposal for the reorganisation of the Capacity Assurance Mechanism in

Table of Relevant Non-Legislative Documents the interconnected system, 29 July 2014. . . . . . . . . . . . . . . . . . . . . . . . . . 16.3.3 Ireland NERA Economic Consulting, The capacity remuneration mechanism in the SEM, Study prepared for Viridian, 4 April 2014. . . . . . . . . . . . . . . . . . . . . . . . . . . 3.4.1 Italy Ministry of Economic Development, National Strategy for Energy, March 2013 . . . 17.4.1 AEEG, Annual Report 2013, 31 March 2013. . . . . . . . . . . . . 17.2.1, 17.2.2, 17.3.2.1 AEEG, Consultation Document No 234/2014 R/EEL, 22 May 2014 . . . . . . 17.3.1, 17.3.2, 17.3.2.1 Terna, Schema di proposta di disciplina del nuovo sistema di remunerazione della capacità produttiva di energia elettrica (Framework Discipline) . . . . . . . . . . . . 17.3.2.1, 17.3.2.3 Netherlands ACM, Visiedocument strategische prioriteiten E&G groothandelsmarkten, No 104195/27. O381, 23 July 2013 . . . . . . . . . . . . . . 18.3.2 Ministry of Economic Affairs, Monitoring Report on the Security of Electricity and Gas Supply (Monitoringsrapportage Leverings- en Voorzieningszekerheid Elektriciteit en Gas 2014), 24 October 2014 . . . . . . . . . . 18.2.2, 18.2.3 Norway Official Norwegian Report, Energi- og kraftbalansen mot 2020, NOU 1998:11 . . . . . . . . . . . . . . . . . . . . . . . 19.2.1 Official Norwegian Report, Hjemfall, NOU 2004:26 . . . . . . . . . . . . . . . . . . . . . . . 19.2.1 Ministry of Petroleum and Energy, Expert group report, Et bedre organisert strømnett, 2013 . . . . . . . . . . . . . . . . . 19.2.2 Regjeringen, Norwegian views on European energy issues, non-paper, 16 May 2013 . . . . . . . . . . . . . . . . . 19.3.2.3 NVE, Elsertifikater: Statusrapport, July 2013 . . . . . . . . . . . . . . . . . . . . 19.3.2.3 Thema Consulting Group, Sertifikatkraft og skatt –oppdatering, THEMA Report 201426, May 2014 . . . . . . . . . . . . . 19.3.2.3

xxix

Poland Ministry of Economy, Report on the Results of the monitoring of the security of electricity supply for period between 1.01.2011 and 31.12.2012, Warsaw 2013. . . . . . . . . . . . . . . . . . . 20.2.2, 20.2.3.1 URE, National Report, July 2014 . . . . . . 20.2.1, 20.2.3, 20.4 Ministry of Economy, Energy Policy of Poland until 2050, Draft, Warsaw, August 2014 . . . . . . . . . . . . . . . . . . . 20.2.3 URE, Report on the investment plans in new generation capacity for years 2014–2028 (Informacja na temat planów inwestycyjnych w nowe moce wytwórcze w latach 2014–2028) Biuletyn URE 4(90), 25 November 2014. . . . . . . . . . 20.2.2, 20.2.3.2 Spain CNMC, Consulta pública sobre el mecanismo de pagos por capacidad, 24 May 2012 . . . . . . . . . . . . . . . . . . . . 21.4 CNMC, Propuesta del mecanismo por el que se establece el servicio de garantía de suministro, 5 December 2012 . . . . . 1.2.3.5 CNMC, Informe Marco sobre la cobertura de la demanda de electricidad y gas para los próximos años, Ref No PA006/12, 17 April 2013 . . . . . . . . . . . . . . . . . . 21.2.3 UK DECC, Electricity Market Reform, Consultation Document, Cm 7983, December 2010 (2010 Consultation Document) . . . . 22.2.3.1, 22.2.3.2, 22.2.3.3, 22.2.3.4, 22.3.1.1 DECC, Planning our electric future: a White paper for secure, affordable and low-carbon electricity, CM 8099, July 2011 (White Paper) . . . . . . . . . 1.2.3.1, 22.2.3.3, 22.3.1.2, 22.4.1 DECC, Electricity Market Reform—Capacity Market Impact Assessment, IA No DECC0103, 27 November 2012 (November 2012 IA). . . . 22.2.3.2, 22.2.3.3 Pöyry, Analysis of the correlation of stress periods in the electricity markets in GB and its interconnected systems, Report to OFGEM, March 2013 . . . . . . . . . . 3.3.1.1, 3.4.1, 3.4.2

xxx

Table of Relevant Non-Legislative Documents

DECC, Capacity Market Strawman v11, June 2013. . . . . . . . . . . . . . . . . . . . . . . . 22.3.1.4 DECC, Electricity Market Reform: Capacity Market—Detailed Design Proposals, Cm 8637, June 2013 (Detailed Design Proposals) . . . . . . . . . . . . . . .3.4.1, 22.3.1.4 London Economics, The Value of Lost Load (VOLL) for Electricity in Great Britain— Final Report for OFGEM and DECC, July 2013 . . . . . . . . . . . . . . . . . 8.2, 22.3.3.1 Charles River Associates, Capacity market gaming and consistency assessment— final report, CRA Project No D1898500, prepared for DECC, September 2013 (CRA Report) . . . . . 10.5.1.2, 10.5.1.3, 10.5.1.4, 22.3.4.2 DECC, Electricity Market Reform: Consultation on Proposals for Implementation (document ref ISBN 9780101870627)

10 October 2013 . . . . . . . . 22.3.1.4, 22.3.3, 22.3.4.2 DECC, Electricity Market Reform—Capacity Market Impact Assessment, 24 October 2013 (October 2013 IA) . . . . . . . 22.2.3.3, 22.3.3.7, 22.4.2 DECC, Electricity Market Reform Delivery Plan, Policy paper, 19 December 2013 (ERM Delivery Plan). . . . 22.3.1.4, 22.3.3.1 DECC, Implementing electricity market reform (ERM)—finalised policy positions for implementation of EMR, 23 June 2014 (Implementing ERM) . . . . . . . . . . . . . . 7.1, 7.3.1.1, 7.3.2.1, 7.6.2, 22.3.3.2 DECC, Electricity Market Reform—Capacity Market Impact Assessment, 4 September 2014 (September 2014 IA). . . . . . . . . . 22.2.3.3, 22.3.3.3, 22.5

List of Contributors Carlos Batlle is Associate Professor with Comillas Pontifical University’s Institute for Research in Technology (IIT) in Madrid and Visiting Scholar under the Massachusetts Institute of Technology’s MIT Energy Initiative, where he teaches Economics and Regulation of the Electric Power Sector. He is Electricity Advisor of the Florence School of Regulation. He has worked and lectured extensively on the operation, planning, and risk management modelling of electricity generation markets and networks, and more specifically on electric power system regulation in over 20 countries. Charlotte Beaugonin joined the EDF group in 2011 as a regulation and public law expert. Previously she worked as a lawyer, first with the law firm Baker & McKenzie in Paris from 2004 until 2007, then with Piwnica & Molinié, a French law firm specializing in litigation before the French Conseil d’Etat and Cour de Cassation. Charlotte holds a Masters degree in public law (Paris I - La Sorbonne) and a Masters degree in litigation law (Paris V - René Descartes). Daniel Crevel-Sander joined the EDF Group in 2010, as a competition law expert after having started his career with the law firm Bredin Prat in Paris. In 2014, he became a Legal Manager at the Belgian subsidiary EDF Luminus, and now specializes in energy law. Daniel studied German and French law at the Universität zu Köln, the Université Paris 1 Panthéon-Sorbonne, and the Université Paris 2 Panthéon-Assas. He also holds a postgraduate Diploma in Economics for Competition Law from the King’s College, London. Dominique Finon is Directeur de Recherche at CNRS (Centre National de la Recherche Scientifique), Scientific Councellor of the Chair European Electricity of Paris Dauphine University, and Senior Research Fellow at CIRED (International Research Center on Environement and Development). His research deals with the long term public policies in the energy sector and the power markets. Jean-Michel Glachant is Director of the Loyola de Palacio Energy Policy Programme and the Florence School of Regulation at the European University Institute. He has been advisor of DG Energy and Transport, DG Competition, and DG Research at the European Commission and of the French Energy Regulatory Commission (CRE), and coordinator or scientific advisor of several European research projects (SESSA, CESSA, Reliance, EUDEEP, RefGov). Jean-Michel is a research partner in the CEEPR at MIT (USA), the EPRG at Cambridge University, the EEI at the University of Leuven, and the GIS Larsen at University Paris Sud. He holds a PhD degree in Economics (La Sorbonne, France). After having been employed by the industry and private sector he became assistant professor at La Sorbonne (1981), Associate Professor, and finally full professor (1999). Martin Godfried is an economist with more than twelve years’ experience in regulation, anti-trust, and mergers in energy markets. In 2011 he joined the Agency for the Cooperation of Energy Regulators as Team Leader Policy and Market Monitoring. Previously he worked for six years at the European Commission (EC) in DG Competition where he

xxxii

List of Contributors

drafted several chapters of the Final Report from the Energy Sector Inquiry and a number of Commission Decisions in energy antitrust cases. Before that he worked at the Dutch energy regulator, the Dutch competition authority, and the OECD in Paris. René H. Gonne is Partner at Dechert LLP, Brussels office. He practises in the field of corporate transactions, such as M&A, restructuring, joint ventures and shareholders’ agreements, corporate governance matters, and related litigation. René has also developed expertise in the energy sector, covering corporate, regulatory, competition, and climate change issues, and has been recommended as a leading lawyer for corporate, M&A, and regulatory matters (energy) in Chambers Europe, Chambers Global, Legal Experts EMEA as well as Legal500 EMEA. René is a member of the International Bar Association (IBA) and the Brussels Bar. He holds a JD degree from Katholieke Universiteit Leuven (1979). Francisco Enrique González-Díaz is Partner at Cleary Gottlieb Steen & Hamilton LLP, Brussels office. He specializes in European and Spanish competition law. Prior to joining the firm in 2003, he held a number of positions within the European institution: Between 1998 and 2003, he headed one of the enforcement units of the European Commission’s Merger Task Force. Before that, he clerked at the General Court in Luxembourg (1996–1998), and was a member of the European Commission’s Legal Service (1990–1996). He has written extensively on EU law matters, particularly in the field of antitrust/competition law. He was a lecturer in Private International Law at the University of Alicante from 1985 to 1989. He also serves on the advisory/editorial boards of a number of legal publications. Enrique has degrees in Law from the University of Granada, the Free University of Brussels, and Harvard University, as well as a degree in Economics from the Open University. He is a member of the Bars in Madrid and Brussels. Iñigo del Guayo has been Professor of Administrative Law at the University of Almería, Spain. He is also a Member of the Academic Advisory Group of the International Bar Association (Section on Energy, Environment, Resources and Infrastructure Law SEERIL). He has published widely in the areas of Public Economic Law, Energy Law, and Regulation. He edited and co-authored Energy Law in Europe, 3rd edn (Oxford University Press, 2015). Iñigo holds a PhD and an LLM from the University of Navarre, Spain. Leigh Hancher is Professor of European Law at Tilburg University, and part-time Professor at the Florence School of Regulation. She is also Of Counsel at the Amsterdam office of Allen & Overy LLP. She is a well-known EU law expert and has counselled firms in a broad range of procedures. She is the author of numerous titles, including EU State Aids, 4th edn (Sweet & Maxwell 2012) and EU Competition and Internal Market Law in the Healthcare Sector (with Wolf Sauter, OUP 2012). Adrien de Hauteclocque is Référendaire at the Court of Justice of the European Union and an Adviser of the Florence School of Regulation and the Loyola de Palacio Chair at the European University Institute. He is also a visiting lecturer at École Nationale d’Administration (France). His research interests include EU antitrust and state aid law, competition policy in network industries and the law & economics of energy regulation. Before moving to the EU institutions, Adrien pursued his academic career

List of Contributors

xxxiii

at the University of Manchester and at the European University Institute where he founded the ‘EU Energy Law & Policy’ Area of the Florence School of Regulation. He is a regular speaker in international conferences and has published numerous books, academic articles and book chapters. Adrien obtained his PhD in Law from the University of Manchester (UK) and holds an MSc in Management from EM Lyon (France) and an MSc in Economic Policy from Strathclyde University (UK). Arthur Henriot was Research Fellow at Florence School of Regulation before joining RTE as a tariff analyst in January 2015. His research interests include energy economics and policy, with a special emphasis on the organization of electricity markets. He has published articles on issues related to large-scale integration of intermittent renewable energy sources in different peer-reviewed international journals, including the Energy Journal, Utilities Policies, Energy Economics, and Energy Policy. Arthur holds a PhD in Economics from Paris-XI University, an MSc in Energy Policy from Imperial College of London, and a Degree in Engineering from the French École Polytechnique. Harald Kröpfl is currently working as Legal Advisor for the Balance Group Coordinators in electricity and gas as well as for the Energy Exchange Austria. In his previous business career he worked for several international law firms specializing in the areas of energy and corporate law. Paolo Mastropietro is an environmental engineer specializing in energy and sustainability and he is pursuing his PhD on electricity market regulation. He joined the Institute for Research in Technology at Comillas Pontifical University after three years in the energy sector. Paolo’s research focuses on the design elements of capacity remuneration mechanisms and their implementation in a regional market context. Antonis Metaxas lectures EU Law at the University of Athens. He is Visiting Professor of Energy Law, IHU, Chairman of the Hellenic Energy Regulation Institute and Managing Partner at Metaxas & Associates Law Firm. Antonis has given numerous speeches and public lectures and has published extensively in the field of energy and European State aid law in Greece and abroad. Jens Naas-Bibow heads Thommessen Law Firm’s practice area of renewable energy. He deals with matters concerning energy law, environmental law, and litigation. He has particular experience with energy licences, European and Norwegian regulation of the energy sector, economic regulation of energy distribution and transmission, district heating, rights of way for infrastructure, expropriation, appraisal cases, and energy law-related litigation. Jens is ranked as a leading lawyer within his field in national and international rankings. Peter Oliver is a former Legal Adviser to the European Commission. He is Visiting Professor at the Université Libre de Bruxelles and a barrister at Monckton Chambers. He has written over seventy publications on EU law, his best known work being Free Movement of Goods in the European Union, 5th edn (Hart Publishing, 2010), written with the aid of six contributors. Ignacio J. Pérez-Arriaga has been Visiting Professor at MIT since 2008 and is currently Professor and Director of the BP Chair on Sustainable Development at the Comillas

xxxiv

List of Contributors

University. He is also Director of Energy Training at the Florence School of Regulation and was formerly Electricity Commissioner in Spain and Ireland. He is a Member of the Board of Appeal of ACER. Ignacio holds an MS and a PhD in Electrical Engineering from MIT, and Electrical Engineering from the Comillas University. Jens Perner is Associate Director at Frontier Economics. He has 18 years’ experience in the energy sector and regularly advises clients on market design, regulatory, competition, and strategic issues. He has led a number of projects on regulation and capacity mechanisms, e.g. recently for the German and Dutch Ministries of Economic Affairs. Jens was previously a Researcher at the Institute of Energy Economics (EWI) at the University of Cologne. He holds a Doctorate in Economics from Cologne University, and a degree in Economics from Hannover University. Jens has frequently published on market and regulatory designs in academic and industry journals. Alberto Pototschnig is the first Director of the European Agency for the Cooperation of Energy Regulators (ACER). Before joining the Agency, he was Partner at Mercados EMI, a Madrid-based international consultancy specializing in the energy sector, where he served as CEO and Deputy Chairman. He previously also served as the first CEO of the Italian Electricity Market Operator and Director of Electricity Regulation at the Italian Energy Regulatory Authority. Since 2004, Alberto has been Adviser at the Florence School of Regulation, where he regularly teaches on energy regulation and market design. Kai Uwe Pritzsche is Partner at Linklaters LLP, based in Berlin. He specializes in M&A transactions, privatizations, and joint ventures, in particular in the energy industry. Kai practised law in Cologne, New York, and Berlin. Since the 1990s he has advised clients in numerous transactions and disputes in the energy industry. Kai is part of the Linklaters global energy sector leadership. He is President of the Board of the Humboldt European Law School Foundation, and a member of the German Institution for Arbitration (DIS). He is Co-editor of the German energy law journal EnWZ and a Member of the Advisory Board of the energy law journal RdE. Kai holds a PhD in Law from the University of Cologne, and an LLM from the University of California, Berkeley School of Law. Katharina Reinhardt is Research Assistant at Linklaters LLP, Berlin office, and a doctoral candidate at the Humboldt-University of Berlin. She studied law and business administration in Berlin and was a Visiting Research Scholar at the University of California, Berkeley. Christoph Riechmann is Director in the Energy Practice of Frontier Economics. He provides market design, regulation, competition, valuation, strategy, and dispute support advice to clients in Europe. He has 20 years’ consulting experience in the energy sector and has recently led several studies on aspects of energy market design at the EU level and in different EU member states. Christoph was previously a researcher at the Institute of Energy Economics (EWI) at the University of Cologne. He holds a Doctorate in Economics (Dr. rer. pol.) from Cologne University, an MSc (Econ) from Glasgow University/Scottish Postdoctoral Programme, and Dipl-Ök. from Justus-Liebig-University, Gieβen (Germany). Christoph has frequently published on

List of Contributors

xxxv

market and regulatory designs in academic and industry journals as well as in several handbooks. Pablo Rodilla is Researcher with Comillas Pontifical University’s Institute for Research in Technology (IIT) in Madrid. He also lectures at the ICAI School of Engineering of Energy Economics, Microeconomics and Electric Power Systems Regulation. Pablo has provided consultant services for governments, international organizations, industrial associations, and utilities in several countries. He has published over twenty papers in international journals and conference proceedings. Fabien Roques is Associate Professor at University Paris Dauphine and Senior Vice President with the economics consultancy Compass Lexecon. His expertise spans European power, gas, and carbon dioxide (CO2) markets, with a specific focus on market design and regulation, strategy, and asset valuation issues. Fabien has a thorough expertise of European capacity mechanisms, and has been involved in the past decade as a consultant in the design and impact assessments of capacity mechanisms implemented in Belgium, France, Greece, Italy, Ireland, Portugal, Spain, and the UK. Fabien is a regular contributor to academic and professional journals on economic issues related to the energy industry. Previous roles included leading the IHS CERA European power and carbon research and consulting team, and contributing to the World Energy Outlook as a Senior Economist with the International Energy Agency. Fabien holds an MSc in engineering from the Ecole Centrale Lyon and a PhD in Economics from the University of Cambridge. Małgorzata Sadowska is Research Fellow at the European University Institute, working on the law and economics of energy regulation and competition policy. Since September 2013 she has coordinated the activities of the Energy Law & Policy Area at the Florence School of Regulation, Robert Schuman Centre for Advanced Studies. Prior to joining FSR, she was Researcher at Tilburg Law and Economics Center within a project involving capacity remuneration mechanisms. Małgorzata holds a PhD in Law and Economics (EDLE, 2013) and Masters degrees in Law (University of Gdańsk, 2006) and European Studies (University of Hamburg, 2008). Francesco Maria Salerno is a Senior Attorney at Cleary Gottlieb Steen & Hamilton LLP, based in Brussels. He has extensive experience advising clients in the energy sector as well as in the telecoms and media sector. He received a Master degree in regulation in 1997 and a PhD from the London School of Economics in 2009 with a comparative study of institutional change in the governance of the telecommunications sector in Italy and Great Britain. Francesco is a member of the Bars in Catania, Brussels, and Madrid. Thomas Starlinger is a founding Partner of Starlinger Mayer Attorneys at Law, Vienna, Austria. He regularly advises national and international clients on all aspects of energy law. His expertise covers a wide range of regulatory and industry questions, price revisions and arbitration, as well as proceedings before national regulatory authorities and the European Commission. He also specializes in transactions and public law.

xxxvi

List of Contributors

Wim Vandenberghe is a Brussels-based lawyer, specializing in energy law. He also has extensive experience in general EU law and competition law. He lectures at King’s College London (UK) and KU Leuven (Belgium). Charles Verhaeghe is a Senior Economist at the Compass Lexecon practice in Paris, where he led several projects on market design in Europe. He has about ten years’ experience in the energy sector and a solid background in energy economics and regulation. Before that, Charles worked for CRE, the French energy regulator, where he was a head of department in charge of power market design and cross-border trade. In this context, Charles was active at the national and European level as he chaired several working groups and projects with other regulators for the Agency for the Cooperation of European Energy Regulators (ACER), especially in relation to security of supply and capacity mechanisms. Catherine Ramstad Wenger is Associate at Thommessen Law Firm’s renewable energy department. She deals with matters concerning energy law, environmental law and climate law. She has particular experience with the European and Norwegian regulation of the energy sector, licensing processes and legal due diligence in relation to M&A and restructuring of renewable energy companies. She is currently on a leave of absence to be part of the Norwegian delegation negotiating the international climate agreement. Bert Willems is Associate Professor in Economics at Tilburg University and holds a PhD in economics and a Masters degree in engineering from KU Leuven. His main research interests are market design, competition policy, and economic regulation. He has studied risk management, congestion pricing, and contracting in the electricity sector. Bert has worked at the UCEI Berkeley, the EUI Florence, and TU Dresden, and is Vice-chair of the Benelux Association for Energy Economics (BAEE). Peter Willis is Co-head of Bird & Bird's Competition & EU team, based in London. The Chambers and Partners directory 2014 describes him as having ‘impressive knowledge of the energy sector’. He provides no-nonsense advice on the application of EU and national competition and regulatory rules. He advises on investigations by competition authorities and regulators, cartel leniency applications, sector enquiries, complaints and merger filings. Clients rely on his in-depth understanding of the interaction between competition rules and the complex technical, commercial and regulatory arrangements governing the operation of energy networks and markets. He is well-known for his expertise in relation to the third EU energy liberalization package, REMIT and other sector rules. An increasing part of his work involves advising on disputes between businesses in which competition and regulatory issues arise, including competition follow-on damages claims. Peter is a regular speaker on competition and regulatory issues at conferences around Europe, and is often asked to explain complex competition and regulatory issues in the press and media. He is the author of well-received books and articles on EU and UK competition law and is a member of the Advisory Board of the Competition Law Journal. Marinus Winters is Counsel at Allen & Overy LLP, Amsterdam office. He has in-depth experience of advising on energy law, competition law, and the regulatory aspects of the

List of Contributors

xxxvii

electricity and gas markets. Marinus has acted for several energy and petrochemical companies in civil and administrative procedures relating to almost every aspect of the energy sector, such as tariff regulation for grid operators, access to the grid, unbundling issues, stranded costs, and CO2 emission allowances. In addition, he often advises on the regulatory aspects of the construction of power plants, gas storages, and (offshore) wind farms.

List of Abbreviations ACER ACM AEEG AG APG ARENH BEE BMWi BNetzA CACM CACs CAM CATs CCGT CDC CEE CEE FBMC CEER CfDs CHP CJEU CNE CNMC CONE CRE CREG CSMEM CWE DECC DG DSOs DSR EEA EEAG 2014–2020 EEC EEG EEU EEX EMR ENSO ENTSO-E EnWG EU

European Agency for Cooperation of Energy Regulators Autoriteit Consument en Mark Autorità per l’Energia Elettrica e il Gas Advocate-General Austrian Power Grid AG Regulated Access to Incumbent Nuclear Electricity German Association for Renewable Energies German Federal Ministry of Economic Affairs and Energy Federal Network Agency for Electricity, Gas, Telecommunications, Post and Railway (Bundesnetzagentur) Capacity Allocation and Congestion Management Capacity Availability Contracts Capacity Assurance Mechanism Capacity Availability Tickets Combined Cycle Gas Turbine Caisse des Dépôts et Consignations Central and Eastern Europe Central East Europe Flow-Based Market Coupling project Council of European Energy Regulators Contracts for Difference Combined Heat and Power Court of Justice of the European Union Comisión Nacional de la Energía National Commission for the Markets and Competition Cost of new entry Commission de Régulation de l’Énergie Regulatory Commission for Electricity and Gas Comité de Seguimiento del Mercado Mayorista de Energía Eléctrica Central and Western Europe Department of Energy and Climate Change Directorate-General Distribution System Operators Demand Side Response European Economic Area Energy and Environmental Guidelines 2014–2020 European Economic Community Renewable Energies Act (Erneuerbare-Energien-Gesetz) Expected energy unserved European Energy Exchange Electricity Market Reform El Niño Southern Oscillation European Network of Transmission System Operators for Electricity Energy Industry Act (Energiewirtschaftsgesetz) European Union

xl EU ETS EXAA FADE FCO FERC FG FSR GB GBER 2014 GME GRTN GSE GW HHI IEA ISO ISO-NE ITRE KPG LOLE MWh MVA MYP NEPs NC NOME NRAs NVE NWE OCGTs OEF OEF OFGEM OTC PCR PJM PPC PSO PTRs PUN PV REMIT RES ResKV RIs RTE

List of Abbreviations European Emission Trading Scheme Energy Exchange Austria Fondo de Amortizacion del Deficit Electrico German Federal Cartel Office US Federal Energy Regulatory Commission Framework Guidelines Florence School of Regulation Great Britain General Block Exemption Regulation 2014 Gestore Mercati Energetici Gestore della Rete di Trasmissione Nazionale Gestore dei Servizi Elettrici gigawatt Herfindahl-Hirschman Index International Energy Agency Independent System Operator Independent System Operator New England European Parliament’s Industry, Research and Energy (ITRE) Committee KWK-Punkte Gesetz Loss-of-Load Expectation megawatt hour Megavolt Ampere Multi-year Plan National Energy Plans Network Codes New Organisation of the Electricity Market (Nouvelle Organisation du Marché de l’Electricité) National Regulatory Authorities Norwegian Water Resources and Energy Directorate North-Western European Open Cycle Gas Turbines Firm Energy Obligation (OEF) Obligación de Energía Firme (Firm Energy Obligation) Office of Gas and Electricity Markets Over-the-Counter Price Coupling of Regions Pennsylvania-New Jersey-Maryland Interconnection Public Power Corporation SA public service obligation Physical Transmission Rights prezzo unico nazionale Photovoltaic Regulation on wholesale energy market integrity and transparency Renewable Generation Sources Ordinance on Reserve Power Plants (Reservekraftwerksverordnung) Regional Initiatives Réseau de Transport d’Électricité

List of Abbreviations SAM SEM SEP SGEI SGI SRMC STOR SWD TEU TFEU TSO TTR TYNDP UK URE VOLL

State Aid Modernization exercise Irish Single Electricity Market Samenwerkende Elektriciteits Productiebedrijven NV Service of General Economic Interest Services of General Interest Short Run Marginal Cost Short-Term Operating Reserve Staff Working Document Treaty of the European Union Treaty on the Functioning of the European Union Transmission System Operator Tender for Targeted Resource Ten-Year Network Development Plan United Kingdom Energy Regulatory Office (Urząd Regulacji Energetyki) Value of Lost Load

xli

PART I POLICY

1 EU Policy on Capacity Mechanisms Francisco Enrique González-Díaz1

1.1 Why capacity mechanisms? The missing money problem and RES capacity In 1996, the EU embarked upon the liberalization of the electricity market.2 In 2003 and 2009, two waves of legislation followed,3 and 2014 was declared the year of completion of the internal energy market.4 This ambitious project has never formally included market design. However, a process of spontaneous convergence has led all Member States to adopt a pool market model for wholesale trading. In the pool, producers and importers sell their output through a bidding process to suppliers, traders, and large industrial customers. Bids are arranged in ascending order, according to the marginal costs of generating units (the so-called ‘merit order’), and the price is set at the level of the marginal cost of the most expensive unit dispatched in order to meet demand. However, this model—which is often described as the ‘energyonly’ market—suffers from a notable failure: demand and supply do not necessarily meet at times of extreme scarcity, ie there is no guarantee of the market clearing. Two factors commonly explain this failure. First, demand is inflexible. In other commodities markets, if supply is scarce, the price would rise and demand would shrink until the market clears. By contrast, in electricity markets, customers are often unable to reduce demand. As a consequence, there may be situations where, even if all available generators produce as much electricity as they can, they are still unable to meet demand. This leads to energy

1 The author would like to thank Francesco Maria Salerno and Tomas Pavelka for their substantial contribution. All errors remain those of the author. 2 Directive 96/92/EC of the European Parliament and of the Council of 19 December 1996 concerning common rules for the internal market in electricity [1997] OJ L 27 (1996 Electricity Directive). 3 Second Energy Package, in particular, Directive 2003/54/EC of the European Parliament and of the Council of 26 June 2003 concerning common rules for the internal market in electricity and repealing Directive 96/92/EC [2003] OJ L 176/37 (2003 Electricity Directive) and Regulation (EC) No 1228/2003 of the European Parliament and of the Council of 26 June 2003 on conditions for access to the network for cross-border exchanges in electricity [2003] OJ L 176/1; Third Energy Package, in particular, Directive 2009/72/EC of the European Parliament and of the Council of 13 July 2009 concerning common rules for the internal market in electricity and repealing Directive 2003/54/EC [2009] OJ L 211/55 (2009 Electricity Directive) and Regulation (EC) No 714/2009 of the European Parliament and of the Council of 13 July 2009 on conditions for access to the network for cross-border exchanges in electricity and repealing Regulation (EC) No 1228/2003 [2009] OJ L 211/15 (2009 Crossborder Regulation). 4 Communication from the Commission to the European Parliament, the Council, the European Economic and Social Committee, and the Committee of the Regions: Making the internal energy market work, of 15 November 2012, COM(2012) 663 final (2012 Internal Energy Market Communication).

4

EU Policy on Capacity Mechanisms

rationing (rolling blackouts). Because supply and demand do not meet, the market does not send any meaningful price signal at times of extreme scarcity.5 Second, prices may be capped. Assuming there is enough capacity to meet demand even at peak times, the energy price will rise to reflect the marginal operating cost of the most expensive unit dispatched. When the margin between available capacity and peak demand tightens, electricity prices will rise to the point of reflecting a scarcity premium.6 Prices may then reach extremely high values, potentially up to the value of lost load (VOLL), when consumers prefer not to consume energy.7 For reasons of sociopolitical acceptability and functioning of the energy-only market, prices may be capped.8 In these cases, supply will meet demand (in the short term), but the capped price might not be enough to cover the fixed costs of the producing power plants, which in turn may deprive the market of the necessary investment in new or replacement capacity.9 Thus, the energy-only market may not provide the price signal which would guarantee an adequate level of generation. This situation is commonly described as the missing money problem, because of the failure to provide high enough returns to maintain the level of capacity adequate to meet demand.10 Capacity mechanisms address this failure by providing the payments deemed necessary to support an appropriate level of generation adequacy, in addition to the revenues from the

5 See Peter Cramton, Alex Ockenfels, and Steven Stoft, ‘Capacity Market Fundamentals’, Economics of Energy & Environmental Policy 2 (2) (September 2013) p 28. See also Laurens J. de Vries and Rudi A. Hakvoort, ‘The question of generation adequacy in liberalized electricity markets’ (CEPS INDES Working Papers 2004) p 2. 6 For a graphical illustration, see Figure 4.1. 7 VOLL is defined as the value attributed by consumers to unsupplied energy. Therefore, it represents the maximum price that consumers are willing to pay to be supplied with energy and at that price they are indifferent between, on the one hand, being supplied and paying the price and, on the other hand, not being supplied (and paying nothing). VOLL is typically quite high (eg several thousand euros per megawatt hour [MWh]). As noted in a study by London Economics, ‘[a]ccurately estimating VOLL for a given region and a specific type of outage is a challenging undertaking as VOLL depends on multiple factors such as the type of customer affected, regional economic conditions and demographics, time and duration of outage, and other specific traits of an outage.’ See London Economic International LLC, Value of lost load literature review and macroeconomic analysis (Briefing paper prepared for the Electric Reliability Council of Texas— ERCOT, 17 June 2013) p 6. 8 See eg Fabien A. Roques and Nicos S. Savva, ‘Price cap regulation and investment incentives under demand uncertainty’ (Cambridge CWPE and EPRG Working Paper, May 2006) p 2: ‘The use of such price cap in electricity markets is justified theoretically by physical and engineering constraints, which might prevent the market to clear in times of scarcity. The theory of “value of lost load pricing” states that in a perfectly competitive market, a price cap set at the Value of Lost Load (VOLL) results in a socially optimal level of investment in generating capacity, with an optimal duration of power shortages.’ 9 Paul L. Joskow, ‘Capacity payments in imperfect electricity markets: need and design’ (2008) Utilities Policy 16 (3), 159–70, 7–8: ‘policymakers have not been shy about ex post adjustments in electricity market designs and residual regulatory mechanisms, sometimes motivated by a desire to hold up existing generators opportunistically.’ 10 Kathleen Spees, Smaule A. Newell, and Johannes P. Pfeifenberger: ‘Capacity Markets—Lessons Learned from the First Decade’ (2013) Economics of Energy & Environmental Policy 2 (2), 2: ‘Regardless of their theoretical economic benefits, energy-only markets are untenable if policy makers are not able to tolerate potential public backlash in response to price spikes or the rolling blackouts associated with loadshed events. If an energy-only market does not provide returns high enough to maintain the planning reserve marking that regulators desire, the market is said to suffer from a missing money problem.’

1.1 Why capacity mechanisms? Price €/kWh

5

Demand

Supply w/o RES Supply with RES

Price decrease

RES Supply

Quantity (kWh)

Figure 1.1 RES supply and impact on price Source: Author’s own illustration.

energy-only market.11 Against this background, the integration of large stocks of capacity from renewable generation sources (RES) contributes to the generation adequacy problem in two ways. First, RES capacity exacerbates the traditional missing money problem. Second, RES capacity contributes to another market failure. These two points are discussed in the two following paragraphs. First of all, RES production is characterized by very low marginal costs and when it sets the price in the energy-only market, the price level may reach zero or a level close to zero. As a result, all other generators are unable to cover their costs. When demand is met by RES capacity acting together with other technologies, the result is that the supply curve moves to the right (see Figure 1.1). Some conventional plants are therefore pushed out of the merit order and all suffer from low utilization rates. As a result, there is a strong incentive for them to exit the market.12 The case of Bavaria provides a warning example of how highly efficient combined cycle gas turbine (CCGT) power plants were driven out of the market when working only 400 hours in the first three months of their activity. In order to be profitable, CCGT plants would need to operate about 5000 hours per year. In that respect, E.ON explained that while gas prices remained relatively high, the drop in demand for electricity during 2012–2013 resulted in prices dropping as low as 30 per cent, thus

11

The Council of European Energy Regulators’ (CEER) Response of 7 February 2013 to the Commission’s consultation on generation adequacy (discussed in section 1.4.5) downloadable from the Commission’s website (n 80). CEER has admitted that ‘there might be some market flaws [in the energy only markets] that lead to an economical suboptimal level of generation capacity and may necessitate specific policy measures to ensure an adequate level of security of supply.’ See also William W. Hogan, ‘Electricity scarcity pricing through operating reserves: an ERCOT window of opportunity’ (November 2012); Peter Cramton and Steven Stoft, ‘Forward reliability markets: less risk, less market power, more efficiency’ (2008) Utilities Policy 16 (3), 194–201; and Joskow (n 9). 12 David Buchan, ‘Why Europe’s energy and climate policies are coming apart’ Oxford Institute for Energy Studies, SP 28 (July 2013) p 19: ‘[T]he intermittency of renewables makes all other energy sources in the marketplace intermittent too. This is bad business for the owners of gas and coal plants.’

6

EU Policy on Capacity Mechanisms

making the operation of gas power plants unsustainable.13 Also, in the United Kingdom (UK), in addition to oil and coal power stations, gas and diesel plants are facing the dilemma of being closed or mothballed.14 The Office of Gas and Electricity Markets (OFGEM), the British energy regulator, warned that the estimated risk of blackouts in 2015 will rise to one in four, due to mothballing of CCGT plants and low new investments. The reserve capacity margin is presumed to drop from 14 per cent in 2013 to only 4 per cent in 2015/16.15 Further mothballing of gas plants is imminent in Slovakia, France, and elsewhere in the EU.16 Altogether, closures and mothballing of CCGT plants resulted in around €6 billion write-downs on gas-fired power investment in Europe over the course of 2013.17 Given that their marginal costs of operation are extremely low, the distortions caused by RES are rather insensitive to the on-going level of public support. Once a RES plant is installed, it will remain in operation for its entire useful life, approximately twenty years. During this time, the RES capacity will thus continue to produce the effects described earlier, regardless of the price in the energy-only market and of the level of subsides which may be designed to top up that market price. The previously mentioned market distortions have thus become a long-standing feature of systems with high RES penetration. The second point is that RES capacity creates a new market failure. This is because neither wind nor sun can provide a source of firm energy, so a high penetration of intermittent RES in electricity systems must go hand in hand with the availability of suitable back-up capacity. This back-up capacity is normally provided by thermal units. Thus, a new—paradoxical—market failure arises. The capacity which is pushed out of the market by the increasing share of RES is precisely the capacity that can guarantee the secure integration of RES in the electricity system. In conclusion, for the foregoing reasons, in systems characterized by the presence of a large amount of RES, capacity mechanisms are considered to address at least two market failures. The first is the traditional ‘missing money’ problem, where the energy-only market may not provide sufficient revenues to justify investments that would guarantee the desired level of generation adequacy. The second is the need for enough flexible capacity to act as a back-up for RES electricity generation. Capacity mechanisms thus bridge the gap between both environmental and security of supply 13 Reuters citing E.ON chief executive Johannes Teyssen, in ‘Weak demand may force E.ON to mothball new power stations’ (13 March 2013), available at http://www.reuters.com/article/2013/03/13/eon-gaspower-demand-idUSL6N0C5AKD20130313, accessed 1 February 2015. 14 Emily Gosden, ‘Stop picking on UK companies, says SSE boss Ian Marchant’, the Telegraph (22 May 2013), available at http://www.telegraph.co.uk/finance/newsbysector/energy/10074698/Stop-picking-onUK-companies-says-SSE-boss-Ian-Marchant.html, accessed 1 February 2015. 15 The Office of Gas and Electricity Markets (OFGEM), ‘Projected tightening of electricity supplies reinforces the need for energy reforms to encourage investment’ (Press release, 5 October 2012), available at https://www.ofgem.gov.uk/ofgem-publications/76253/20121005capacitypressrelease.pdf, accessed 1 February 2015. 16 See, for example, ‘E.ON considers mothballing Slovak gas power plant’, in Budapest Business Journal, 9 May 2013; ‘EDF plans to mothball two Belgian gas-fired plants’, in Gas to Power Journal, 13 December 2013, ‘EDF Suez confirms plans to mothball 3 gas-fired power plants in France’, in Gas to Power Journal, 11 April 2013. 17 Ben Caldecott and Jeremy McDaniels, ‘Stranded generation assets: Implications for European capacity mechanisms, energy markets and climate policy’ (Working Paper, Smith School of Enterprise and the Environment, University of Oxford, January 2014) p 7.

1.2 Core features of capacity mechanisms

7

concerns and in fact enable the development of a harmonious environmental policy in the EU.

1.2 Core features of capacity mechanisms Member States have implemented a variety of capacity mechanisms. All these capacity mechanisms are characterized by the presence of certain core elements, which are described in the following sections.

1.2.1 Supply and demand of capacity Capacity means the ability to physically generate electricity, subject to minimum performance requirements. This definition entails two consequences. First of all, there is little, if any, room to introduce a mechanism based exclusively on financial settlements (while this is possible in energy-only markets). This is because the purpose of these markets is to deliver capacity to systems and guarantee generation adequacy. Second, the generator must be able to deliver the capacity to the system, which, in turn, requires action by a transmission system operator (TSO). Thus, the TSO—acting alone or in cooperation with the TSOs of the neighbouring systems (countries)—plays a key role in the functioning of a capacity mechanism for physical reasons. TSOs may also play another important role in capacity mechanisms: that of the central capacity purchaser. On the supply side, capacity can be provided by qualified generators, ie those which comply with certain reliability requirements. Capacity mechanisms are normally open to both new and existing plants. The rationale for remunerating existing plants becomes apparent when taking into account the two previously mentioned market failures. Indeed, as some existing plants need to act as a back-up for intermittent RES, there is a clear need to keep them in the system. Capacity mechanisms can also include demand side resources, capable of being interrupted upon demand. This includes essentially interruptible customers, providing a service to the system which is akin to that of back-up generators. For this reason we refer to ‘capacity providers’ throughout this book whenever we mean both generators and providers of demand side resources. Demand side resources will be discussed in more detail in section 1.2.2 below. On the demand side of capacity mechanisms, there are essentially two models. In the first, the centralized model, demand is pooled and capacity is procured jointly, and the TSO acts as a centralized purchaser. In the second, the decentralized model, all suppliers serving retail customers and large customers are obliged to contract a certain level of capacity linked to their self-assessed supply obligations or future consumption, respectively. For instance, an entity would contract its capacity to cover its consumption three years down the line. With this time frame, generators receive adequate signals as to the future needs of the system and secure future revenues for the initiation of projects. The level of capacity required can be administratively set by the TSO, by the regulator or by aggregating the demand of the entities subject to an obligation to buy capacity. Normally, the price for capacity can be set by a tender where generators either

8

EU Policy on Capacity Mechanisms

bid a scarcity price or a capacity payment. In the former case (so-called scarcity price bidding), capacity providers bid the price that they would need to receive (times the amount of capacity offered) in order to remain in or enter the market. The bids are then aggregated in descending order. The lowest bid sets the remuneration for all generators. They will receive this price (in addition to their remuneration in the energy-only market) for delivering the pre-determined amount of capacity during the hours in which the energy-only market signals scarcity. In the latter case, capacity providers bid for a payment which is independent of the electricity price in the energyonly market. The bids reflect the cost of new entry (if the mechanism aims only at mitigating the ‘traditional’ missing money problem) or the cost of being available (if the mechanisms reward flexible back-up generation). However, sometimes tenders are not possible. For instance, at times of overcapacity, when supply exceeds demand, the price may fall to zero.18 In order to avoid such a situation, the regulator would have to set the capacity price. When it comes to performance incentives, the most basic incentive for capacity providers to deliver their committed capacity is that their failure to do so lowers their payment to the point where total non-performance could result in forfeiting the payment altogether. Generators may also be required to minimize their outage times due to maintenance reasons. Thus, for instance, a capacity mechanism may allow a maximum number of maintenance outages, but provide that for any additional outage time, the payments will be reduced proportionally.

1.2.2 The role of demand side response (DSR) The mirror image of maintaining or increasing existing capacity is lowering or optimizing consumer demand for electricity, that is, demand side response (DSR). The basic premise of DSR is that it may provide a cheaper alternative to conventional generation.19 According to the Commission, demand side resources in the EU could contribute around 60 gigawatt (GW), ie the capacity of approximately 60 nuclear power reactors or 120 middle-size CCGTs.20 However, in the discussion about DSR, it is necessary to distinguish between large industrial customers on the one hand, and smaller consumers and households on the other. Some Member States have already designed schemes to tap into the potential of energy-intensive customers to provide network stability.21 The provision of 18 Steven Stoft, Power System Economics: Designing Markets for Electricity (New York: IEEE Press and Wiley-Interscience, 2002) p 48. See also Sebastian Schwenen, ‘Strategic bidding in multi-unit auctions with capacity constrained bidders: the New York capacity market’ (EUI Working Paper RSCAS 2012/62, November 2012). 19 Roger Levy, Karen Herter and John Wilson, ‘Unlocking the potential for efficiency and demand response through advanced metering’ (Conference Paper, University of Berkeley, June 2004). 20 Commission staff working document, Generation Adequacy in the internal electricity market— guidance on public interventions, 5 November 2013 (Generation Adequacy SWD) p 9. See also Carla Frisch, Martin D. Smith, ‘Electric Utility Demand Side Management: Defining and Evaluating Achievable Potential’, Duke University (May 2008). 21 For example, see OFGEM’s decision of 19 December 2013, empowering the national grid to introduce new balancing services including payments to firms for creating reserves on the demand side, available at

1.2 Core features of capacity mechanisms

9

interruptibility services22 from these customers may be in the form of a dedicated auction or in exchange for an administrative payment. However, the number of customers participating in these schemes is usually limited due to high eligibility requirements in terms of volume of consumption, stability of the service, and the availability of costly equipment to ensure that their production facilities can be disconnected with no harm to the production process. As to small industrial consumers and households, important investments in equipment must be made first to realize their potential contribution to network stability.23 According to a 2011 Commission Communication, despite the heavy investments (€5.5 billion) made in various smart grid projects in recent years, only 10 per cent of EU households were equipped with smart meters as of 2011.24 Industry estimates indicate that more than €50 billion would need to be invested in smart meters by 2020 and another €480 billion to upgrade the rest of the grid system by 2030.25 Moreover, to be effective, demand side response measures by smaller customers need to reach a critical mass. The 2009 Electricity Directive recommended that Member States roll-out 80 per cent of smart metering by 2020,26 generally considered as the critical mass.27 In that respect, Member States were also obliged to perform precise and localized cost-benefit, sensitiveness, and performance tests, including a local assessment of the critical mass to optimize the system.28 All tests were submitted to the Commission, which in 2014 published a benchmarking report.29 Although the report notes that significant progress has been made, the targets will not be attained until 2020 at the earliest. https://www.ofgem.gov.uk/publications-and-updates/national-grid%E2%80%99s-proposed-new-balancingservices-decision-letter, accessed 1 February 2015. 22 The provision of interruptibility services consists of the reduction of power supplies to a customer following an order from the TSO in order to re-balance the electricity network. See for example the Spanish scheme as established by Ministerial Order No IET/2013/2013 of 31 October 2013, Spanish Official Bulletin No 262 of 1 November 2013, available at https://www.boe.es/diario_boe/txt.php?id=BOE-A-2013-11461, accessed 1 February 2015. 23 Cramton, Ockenfels, and Stoft (n 5) p 27: ‘The main problem is a lack of real time meters and billing and other equipment to allow consumers to see and respond to real time prices, resulting in low demand flexibility.’ 24 Commission Communication, Smart Grids: from innovation to deployment, 12 April 2011, p 4. 25 EurActiv.com special report quoting Jan Panek, Head of Unit for Internal Market at DG Energy (27 June 2013), available at http://www.euractiv.com/special-report-building-way-cris/eu-smart-meterroll-lags-ambitio-news-528914, accessed 1 February 2015. 26 Annex I.2 of the 2009 Electricity Directive (n 3) reads: ‘Where roll-out of smart meters is assessed positively, at least 80% of consumers shall be equipped with intelligent metering systems by 2020’. See also Recital 11 of the 2009 Electricity Directive: ‘In order to promote energy efficiency, Member States or, where a Member State has so provided, the regulatory authority shall strongly recommend that electricity undertakings optimize the use of electricity, for example by providing energy management services, developing innovative pricing formulas, or introducing intelligent metering systems or smart grids, where appropriate.’ 27 DECC, Smart meter roll-out for domestic sector, April 2012, p 23. 28 European Commission, Commission Recommendation 2012/148/EU of 9 March 2012 on preparations for the roll-out of smart metering systems [2012] OJ L 73/9. 29 Member States were required to submit their tests to the Commission by September 2013. See Commission Staff Working Document ‘Incorporating demand side flexibility, in particular demand response, in electricity markets’, 5 November 2013, p 12. The Commission published the benchmarking report on smart metering deployment on 17 June 2014, accompanied by the state-of-play of smart metering implementation in the EU and an overview of the cost-benefit analyses conducted by Member States. See Report from the Commission, Benchmarking smart metering deployment in the EU-27 with a focus on electricity (COM(2014) 356 final, 17 June 2014, Brussels).

10

EU Policy on Capacity Mechanisms

In sum, despite having the potential to lower and/or optimize consumer demand for electricity, in the short to medium term, demand side response measures cannot eliminate the need for capacity mechanisms.

1.2.3 Capacity mechanisms in the EU Whilst there are is a variety of capacity mechanism designs, a key difference relates to the competitive or regulated nature of the mechanism, namely whether it is a regulated approach or a market based mechanism that determines the price and/or volume of capacity. Market based mechanisms are commonly referred to as ‘capacity markets’. There are many different ways to classify capacity mechanisms. It is common to group capacity mechanisms into: (a) price-based and volume-based, (b) market-wide and targeted, (c) centralized and decentralized. In a price-based capacity mechanism, policy makers set price and let the investors decide how much they are willing to invest at a given price. To the contrary, in a volume-based mechanism, policy makers decide on the volume of capacity required, allowing market forces to determine its price. Market-wide capacity mechanisms reward all capacities, whereas targeted mechanisms reward only specific plants or technologies.30 In centralized capacity mechanisms contracts are awarded centrally, as opposed to decentralized mechanisms where contracts are awarded through bilateral arrangements. This book refers to the taxonomy provided by the Agency for Cooperation of Energy Regulators (ACER), which defines five different types of capacity mechanisms: strategic reserve, capacity obligation, capacity auction, reliability option and capacity payment (see Figure 1.2).31 Figure 1.3 provides a snapshot of the five types of capacity mechanisms existing in the EU.32 While all result from different combinations of the core features presented earlier, each of these five types can have further variants.

1.2.3.1 Strategic reserve In this mechanism, a central agency (eg the TSO or a governmental authority) decides the amount of capacity needed a number of years in advance. This capacity is then contracted, usually by means of a tender based on the lowest price offers. The plants so contracted are then held in reserve and no longer participate in the energy market. Only when capacity shortfalls arise are they activated according to some pre-defined 30

Some price-based mechanisms can be targeted as well. ACER, Report on capacity remuneration mechanisms and the internal market for electricity, 30 July 2013 (ACER’s Report) p 5. See below, section 2.5 for the main conclusions presented in ACER’s Report. 32 Capacity mechanisms also exists outside the EU and some non-European country examples are provided in sections 1.2.3.1–1.2.3.5. 31

1.2 Core features of capacity mechanisms

11

Capacity Remuneration Mechanisms

Volume based

Targeted

Strategic reserve

Price based

Market-wide

Capacity obligation

Capacity auction

Reliability option

Capacity payment

Figure 1.2 Possible models for capacity mechanisms Source: ACER, Report on capacity remuneration mechanisms and the internal market for electricity, 2013 (ACER’s Report).

criterion, and their output is sold via the general energy market. This trigger criterion can be, for instance, a certain (high) price level in the energy-only market (threshold price).33 As a consequence, strategic reserves are capacities set aside in order to secure supply in certain exceptional circumstances of extreme scarcity. Given this targeted nature, strategic reserves normally have only a small impact on the prices in the energy-only market.34 Economic considerations suggest that particularly old plants will be best suited to act as reserve in this regime.35 Strategic reserve has been implemented in Sweden and Finland, which are hydrodominated and need to ensure enough capacity reserve to meet demand in case of a dry year. In addition, Sweden had to address the lack of sufficient capacity after the decommissioning of a large stock of nuclear generation in the late 90s. Interim solutions similar to strategic reserve are currently in place in Germany, Belgium, and Poland.36 Beyond the EU, strategic reserves have also been used in Australia37 and New Zealand.38

33 When the price in the energy-only market reaches a certain threshold, reserve capacity is activated and becomes now price-setting capacity. Therefore, strategic reserve has the equivalent effect to setting a price cap in the market. 34 Regulatory Commission for Electricity and Gas (CREG), Capacity remuneration mechanism, Study (F)121011-CDC-1182, 11 October 2012, p 11 (CREG’s study on capacity mechanisms). Also, the Commission notes that strategic reserves avoid the ‘wait for tender’ problem while at the same time not affecting the market during normal periods. See Communication from the Commission, Delivering the internal electricity market and making the most of public intervention, C(2013) 7243 final (November 2013 Communication) p 21. 35 Apparently this is the preferred alternative now in Germany. As a reference, it was largely discussed (and criticized) in the public consultations launched by DECC at the start of the Electricity Market Reform (EMR). 36 See Germany (chapter 15), Belgium (chapter 13), and Poland (chapter 20). 37 DECC, Planning our electric future: a White paper for secure, affordable and low-carbon electricity, July 2011, p 200 (White Paper). 38 Leonie Meulman and Nora Méra, ‘Capacity mechanisms in northwest Europe. Between a rock and a hard place?’ (2012) Clingendael International Energy Programme, p 55.

Existing capacity mechanisms SR SR

Some elements of a capacity mechanism in place (types of strategic reserves) Implementation of a capacity mechanism or its revision in progress

CP

CA CA

RO

SR

Capacity mechanism under consideration CO SR – strategic reserve CA – capacity auctions CO – capacity obligations RO – reliability options CP – capacity payments

CP CP

CP CA

CP RO CP CO

Figure 1.3 Capacity mechanisms in Europe, 2015 Source: Author’s own illustration.

1.2 Core features of capacity mechanisms

13

1.2.3.2 Capacity auctions Similar to strategic reserves, capacity auctions are one-off interventions. The total required capacity is set several years in advance of delivery and centrally procured in an auction by an independent body (usually a TSO). Capacity providers bid to receive a capacity payment, which reflects the cost of building new capacity. The lowest offer wins the auction, as the TSO is interested in the lowest possible price for the capacity to be sold by generation companies. The new capacity selected at the auction will participate in the energy-only market. As a consequence, there may be a market distortion if the amount of capacity is such that the new capacity may artificially undercut the existing capacity in the energy-only market. In addition, there is a risk of triggering a ‘wait-for-the-tender’ approach, where investors refrain from responding to the energy-only market price signals for fear of losing the extra remuneration they may reap in capacity auctions. The UK capacity mechanism recently implemented and approved by the Commission is based on centralized auctions.39 Capacity auctions are also discussed in Germany and Poland as a long-term option. Beyond the EU, capacity auctions have also been used in the US, Colombia, Brazil, and Panama.40

1.2.3.3 Capacity obligations This mechanism imposes an obligation on large consumers or suppliers to contract a certain level of capacity linked to their self-assessed future consumption or supply requirements (plus a certain level of reserve margin, established by regulation). The obligation can be fulfilled by owning generation facilities (self-supply) or entering into bilateral agreements with capacity providers. A key design aspect of this system is that suppliers (or consumers) are penalized financially if, in certain situations of capacity scarcity (to be defined by the regulator), they have not procured the required capacity. The expectation is that the threat of such penalties will promote the development of a market for capacity contracts that offer an additional (and relatively stable) revenue stream for plant operators. To avoid penalties, suppliers and consumers need to prove that together with their contracted energy they also bought the associated plant capacity. This can be done through certificates, each of which represents a certain amount of capacity made available by a given capacity provider. If they are standardized, they can be traded in a separate market, where prices are set by supply and demand (certificate market). On the supply side, contracted capacity providers are required to make the contracted capacity available to the market in periods of shortage, defined administratively or by market prices rising above a threshold level.

39

See chapters 14 and 22 on the French and UK capacity mechanisms. It should be noted that there are no real capacity auctions in Latin America. However, there are longterm energy auctions which are discussed in detail in chapter 7. 40

14

EU Policy on Capacity Mechanisms

Greece adopted a capacity mechanism based on capacity obligations in 2005 (not implemented in practice),41 and France will follow suit in 2014.42 Furthermore, this model is strongly favoured by the German energy industry. Going beyond European examples, this model can also be found in the US, where regional pools introduced a form of payment for capacity indirectly through capacity obligations imposed on load serving entities. Such capacity mechanisms exist in the Independent System Operator New England (ISO—NE), New York ISO and the Pennsylvania-New Jersey-Maryland Interconnection (PJM)43 in the District of Columbia.44

1.2.3.4 Reliability options In this model, a counterparty (for instance, TSO or large customers and suppliers) is designated by the regulator to enter into option contracts with capacity providers. Such contracts offer the counterparty the option to procure power at a strike price. The counterparty will make use of this option if electricity spot prices spike in a scarcity situation and exceed the strike price. In other words, the option contract obliges the capacity provider to pay the counterparty the difference between the price in the energyonly market and the strike price, whenever this difference is positive (contracts for difference, CfDs). In exchange for this price guarantee, capacity providers receive an option premium, which is fixed and provides a more stable and predictable stream of revenues (similar to capacity payments). In this regime, capacity providers will continue to participate in the energy market. The strike price, and the total volume of the reliability options that capacity providers must offer, is set through regulation.45 In that respect, these mechanisms differ from other risk-hedging instruments available in the market, where both the strike price and the volume are negotiated in the market.46 Further, reliability options are normally not just financial contracts, but also entail the physical delivery of 41 Grid and Market Operation Code, RAE, 2005. For more details, see chapter 16 on the Greek capacity mechanism. See also Kostis Sakellaris: ‘The Greek capacity adequacy mechanism: design, incentives, strategic behavior and regulatory remedies’ (April 2009) p 1. 42 In France, the so called NOME Law (Nouvelle Organisation du Marché de l’Electricité, New Organisation of the Electricity Market) provides a legal basis for implementing a capacity mechanism in the form of exchange of capacity certificates from 2014 onwards, see eg Anna Creti, Jerome Pouyet, Maria Eugenia Sanin, ‘The Law NOME: Some Implications for the French Electricity Markets’ (CEPREMAP Working Paper 1102, 8 March 2011). See also chapter 14 on France. 43 PJM stands for Pennsylvania-New Jersey-Maryland Interconnection LLC (Mid-Atlantic region power pool). Since its establishment in 1927, PJM has continued to integrate additional utility transmission systems into its operations and currently covers all or parts of Delaware, Illinois, Indiana, Kentucky, Maryland, Michigan, New Jersey, North Carolina, Ohio, Pennsylvania, Tennessee, Virginia, West Virginia, and the District of Columbia. 44 Tomás Gomez San Román (CNE), ‘Capacity markets design with a growing RES penetration’ (presentation at the 1st OMIE International Workshop, 3 July 2012, Madrid), p 6, available at http:// www.cne.es/cgi-bin/BRSCGI.exe?CMD=VEROBJ&MLKOB=734701011919, accessed 1 February 2015. 45 The strike price is normally based on the short run marginal cost of a new entrant using the most efficient technology, often a combined cycle gas turbine (CCGT). As a consequence, the generators will have to reimburse all extra benefits that they would have received from the energy-only market. Linking the capacity mechanism to the revenues in the energy-only market thus ensures that generators receive only the minimum support necessary to go beyond what market forces alone would have achieved. For an example of this approach, see chapter 17 on Italy’s reliability option model. 46 In case of other instruments for hedging price volatility.

1.2 Core features of capacity mechanisms

15

electricity. In particular, when the price in the energy-only market exceeds the strike price, the capacity provider must be available, otherwise it will pay a penalty. In any case, capacity providers are selected in a competitive tender, which will also set the capacity premium. Reliability options act as a cap on the price that generators can earn in the energyonly market. Thus, the generators’ total revenues are derived from two streams: (a) the capacity premium; and (b) the energy-only market, where, however, the revenues are capped at the value of the strike price. As a consequence, too low a strike price may cause generators to demand a high capacity premium to compensate for the shortfall. On the other hand, the mechanism ensures that generators receive exactly the support needed to provide the service. Reliability options are being implemented in Italy, through a central auctioning.47 In addition, this model has been recommended in two studies for the German Federal Ministry of Economic Affairs and Energy (BMWi).48 The UK reform of the system of imbalances may lead also to a market for reliability options.49 Beyond the EU, the US and Colombia have also implemented reliability options.

1.2.3.5 Capacity payments Capacity payments are pre-determined fees set by the regulator and paid to capacity providers. They have been in place for several years in markets that are less well connected in the periphery of Europe such as Spain, Portugal, Ireland, Greece, and Italy. Capacity payments may be determined in various ways, they can be fixed or variable (eg per ‘firm’ megawatt [MW]50 and year or simply per month), and awarded to all or part of the eligible capacity declared or actually available. For instance, the Spanish capacity payments are based on the estimated capacity margin,51 whereas in Ireland, they are based on the cost of new entry estimated by the regulator each year. The plants so rewarded will continue to participate in the energy-only market where they will still benefit from an additional revenue stream. As such, capacity payments complement the energy-only market income, as it is assumed that the latter is insufficient to (a) attract new capacity needed to cover demand; and/or (b) to provide incentives for capacity of a certain kind to remain in the system. The recipients of capacity payments can be selected through a tender. However, for capacity payments designed to remunerate back-up capacity in systems with a high RES penetration, the feasibility of the tender depends on the proportion between the available firm capacity and the RES capacity. In some cases, such as Spain, the amount 47

See chapter 17 on Italy. Energiewirtschaftliche Institut an der Universität zu Köln (EWI), Untersuchungen zu einem zukunftsfähigen Strommarktdesign (2012); Wissenschaftlicher Beirat des BMWi, Langfristige Steuerung der Versorgungssicherheit im Stromsektor (November 2013). Both studies are available in German only. 49 See Commission decision of 23 July 2014 in Case SA.35980 (2014/N-2) United Kingdom Electricity Market Reform—Capacity Market, C(2014) 5083 final, [2014] OJ C/348 (UK capacity mechanism) para 92. 50 ‘Firm’ megawatt [MW] refers to the expected contribution of each plant to the system (the capacity credit in the US jargon). See chapter 7 based on the Latin American experience. 51 See chapter 21. 48

16

EU Policy on Capacity Mechanisms

of RES may be so high that all existing firm capacity is needed. As a consequence, a tender may not be feasible, and capacity payments are paid out to all the providers of firm capacity.52 Spain has used capacity payments since 1997,53 Portugal since 2010, and Ireland and Greece have used this system since 2005.54 Italy recently decided to move to an auction system and reforms are planned in Spain and Greece, which could lead to a change from the current approach relying on administratively set capacity prices toward a more market based approach. Looking beyond the European experience, capacity payments have been introduced in some South American countries including Argentina, Brazil, Chile, and Colombia.55

1.3 The EU approach to capacity mechanisms before the 20/20/20 Package 1.3.1 The 2003 Electricity Directive The provisions relating to the construction of new capacity in the 2003 Electricity Directive56 show that the EU approach to capacity mechanisms before the 20/20/20 Package57 focused on the missing money problem and capacity tenders to attract investment in new capacity. Article 6 of the Directive provides that Member States should have authorization procedures for the construction of new capacity according to certain pre-defined criteria set out therein.58 For the construction of new capacity, therefore, the market must go through the authorization procedure. However, if this procedure fails, ie there is insufficient spontaneous reaction from the market, then, but only then, may Member States organize tenders for new capacity according to Article 7. Indeed, this provision explicitly conditions tenders on the failure of the authorization procedure, and the ensuing risk for security of supply by stating that ‘[the tender] can, however, only be launched if on the basis of the authorization procedure the generating 52 Report from the Spanish Energy Regulator (CNE), Propuesta del mecanismo por el que se establece el servicio de garantía de suministro (5 December 2012) p 22. 53 See chapter 21 for a discussion. 54 See chapter 16 for more detail on the Greek capacity payments. 55 See, for example, Fereidoon P. Sioshansi, Competitive Electricity Markets: Design, Implementation, Performance (Elsevier, 2008) p 336. 56 2003 Electricity Directive (n 3). 57 The climate and energy package (the 20/20/20 Package) is a set of binding legislation which aims to ensure the EU meets its ambitious climate and energy targets for 2020. These targets, known as the ‘20/20/ 20 targets’, set three key objectives for 2020: (a) a 20% reduction in EU greenhouse gas emissions from 1990 levels; (b) raising the share of EU energy consumption produced from RES to 20%; (c) a 20% improvement in the EU’s energy efficiency. More information on the Package is available at http://ec.europa.eu/clima/ policies/package/index_en.htm, accessed 1 February 2015. See also section 1.4. 58 2003 Electricity Directive (n 3) Art 6(1) and 6(2): ‘For the construction of new generating capacity, Member States shall adopt an authorisation procedure, which shall be conducted in accordance with objective, transparent and non-discriminatory criteria. Member States shall lay down the criteria for the grant of authorisations for the construction of generating capacity in their territory. These criteria may relate to: [among others] safety and security of the electricity system, protection of public health and safety, protection of the environment, energy efficiency, the nature of the primary sources, characteristics particular to the applicant, compliance with measures adopted pursuant to Article 3 [public service obligation].’

1.3 EU approach before 20/20/20 Package

17

capacity being built or the energy efficiency/demand side management measures being taken are not sufficient to ensure security of supply.’59

1.3.2 The 2003 Irish CADA case A good illustration of this approach is the 2003 Commission state aid decision on the Irish tender for new capacity.60 In this case, the Irish authorities showed that there was ‘an important capacity shortfall from 2005 onwards.’61 In other words, the energy-only market did not provide the right incentives (revenues) for the construction of new capacity. Under these circumstances, the Irish authorities decided to organize a tender for new capacity. In parallel they implemented a mechanism linking the generators’ remuneration with the price in the energy-only market. Namely, the generators had to make payments to the TSO whenever the electricity spot price exceeded the strike price. The Commission assessed the measure in light of the Altmark judgment,62 delivered in the same year63 and made the following findings. First, the Commission considered that the generators that were going to deliver the capacity could be seen as being subject to a public service obligation (PSO).64 Second, the Commission took the position that, for such a PSO to be validly imposed, Ireland had to show that its intervention was both objectively justified on public service grounds and proportionate. 59

Article 4 of the 1996 Electricity Directive (n 2) did not confer any priority on authorizations over tendering. The Commission justified the change in Directive 2003/54 by adducing the following reasons: ‘launching a tendering procedure constitutes an intervention on the market from the part of the authorities; such a procedure, as is the case with other interventions, distorts the investment signals that exist in the market and could lead to “a wait for the tender to be launched” approach on the part of investors’, see Note of DG Energy & Transport on Directives 2003/54 and 2003/55 on the internal market in electricity and natural gas, 16 January 2004, p 6. 60 Case N 475/2003 Irish CADA [2003] OJ C 34/7. See Case C-280/00 Altmark Trans [2003] ECR I-07747, paras 88–94. See also Case N 143/2004 Ireland, Public Service Obligation—Electricity Supply Board [2005] OJ C/242. In 2004, Ireland adopted a temporary measure for the provision of the reserve capacity. This temporary solution was the imposition on the TSO of a public service obligation consisting mainly of the procurement of temporary generation units. The Commission found that the Altmark case law was not applicable because the fourth condition was not met. However, it authorized the measure as compatible aid based on Article 106(2) TFEU. 61 Irish CADA (n 60) para 5. 62 Altmark (n 60). In its Altmark judgment the Court of Justice of the European Union (hereafter: CJEU, Court) held that public service compensation does not constitute state aid when four cumulative conditions are met: (a) the recipient undertaking must have public service obligations and the obligations must be clearly defined; (b) the parameters for calculating the compensation must be objective, transparent, and established in advance; (c) the compensation cannot exceed that which is necessary to cover all or part of the costs incurred in the discharge of the public service obligations, taking into account the relevant receipts and a reasonable profit; and (d) where the undertaking which is to discharge public service obligations is not chosen pursuant to a public procurement procedure which would allow for the selection of the tenderer capable of providing those services at the least cost to the community, the level of compensation needed must be determined on the basis of an analysis of the costs of a typical well-run company. 63 Altmark (n 60). 64 Public service obligations (PSOs) may be imposed by the public authorities on the body providing a service (airlines, road or rail carriers, energy producers and so on). See Irish CADA (n 60) para 22: ‘The Commission considers that the measures undertaken by the Irish authorities in order to ensure an adequate security of supply must be understood as imposing on generators an obligation of general economic interest which consist in bringing to the Irish electricity grid new electricity reserve generation capacity in order to be sure to be able to meet the electricity demand in the future at any time of the year, including in peak periods.’

18

EU Policy on Capacity Mechanisms

The Commission further divided the ‘objective justification’ test in two limbs. First, the capacity at issue must be ‘reserve’ capacity, not ‘normal’ capacity. In that respect, the Commission defined reserve capacity as: The additional capacity that would not be spontaneously provided by normal market forces but is considered necessary in order to meet peaks of demand. One may indeed wonder whether investors are prepared to invest in peaking capacity to cover the very highest periods of demand or incidents where a large proportion of other generation is not available. It is arguable that such investment might not occur because such events are infrequent and their occurrence is unpredictable. Accordingly there may be a case for governments to provide further measures, in addition to market mechanisms, to ensure adequate capacity is available.

Secondly, in order for the PSO to be objectively justified, the need for new reserve capacity must be clearly and quantitatively demonstrated. On proportionality, the Commission required that, first, the tendered capacity must be within reasonable limits (‘reasonable standards of capacity reserve’) and second, that the need for the reserve capacity must be established taking into account demand side measures and the use of existing interconnectors. Having been satisfied that the above conditions were met, the Commission concluded that ‘the first condition of the Altmark judgment is fulfilled’, ie that the generators were ‘actually required to discharge public service obligations and those obligations have been clearly defined.’ As to the other Altmark conditions, the Commission emphasized that the use of a tender, coupled with the mechanism to retrocede any excess payment, ensured that: (a) the parameters for compensation were set ex ante in an objective manner (second Altmark condition); (b) there was no overcompensation (third Altmark condition); and (c) the provider was selected in a way that minimized costs, that is by a tender (fourth Altmark condition). As a consequence, the Commission concluded that the scheme notified by the Irish authorities fulfilled all four Altmark conditions, and it did not constitute state aid.

1.4 The EU approach to capacity mechanisms after the 20/20/20 Package (and before the November 2013 Communication) In March 2007, the European Council agreed a new EU energy policy (the 20/20/20 Package) which included three main pillars: (a) reducing greenhouse gases emissions in the EU by 20 per cent by 2020; (b) ensuring that 20 per cent of the EU’s overall energy consumption is met by RES by 2020 and (c) reducing energy consumption in a cost efficient manner by 20 per cent by 2020.65 The cornerstone of the second pillar was the 2009 RES Directive,66 which mandated RES targets for Member States. As a consequence of the targets, Member States enacted support systems for RES. The availability of financial support spurred investment in RES capacity. 65

The 20/20/20 Package (n 57). Directive 2009/28 on the promotion of the use of energy from renewable sources and amending and subsequently repealing Directives 2001/77/EC and 2003/30/EC [2009] OJ L 140 (2009 RES Directive). 66

1.4 EU approach after 20/20/20 Package

19

The resulting picture is one in which capacity payments have a dual rationale. First, they continue to be necessary for generation adequacy reasons and to attract investment in new capacity. Second, capacity payments secure generation flexibility. They incentivize thermal capacity to stay in the system and provide the backup service to fully integrate RES into the electricity systems. As this latter rationale became more and more prominent, on 22 May 2013 the European Council called for particular priority to be given to the Commission providing guidance on capacity payments. The Commission met this call with the Communication on Delivering the internal electricity market and making the most of public intervention of 5 November 2013 (the November 2013 Communication, discussed in detail in section 1.5 below), which assesses the main features of public interventions in the electricity market and provides guidance on how they can be designed or respectively adapted in order to increase their effectiveness.67 The Commission accompanied the November 2013 Communication with five Staff Working Documents, one of which deals with Generation Adequacy in the internal electricity market—guidance on public interventions (the Generation Adequacy SWD).68

1.4.1 The 2009 Electricity Directive The provisions of the 2009 Electricity Directive69 on the construction of new capacity remained identical to those of the 2003 Electricity Directive. As a result, the focus on the ‘missing money’ problem and the ensuing remedy based on tendering for new capacity, also remained in place. Two state aid decisions provide a useful illustration of the Commission’s approach under the 2009 Electricity Directive.

1.4.2 The 2010 Latvian case The Latvian authorities70 considered that the Baltic region would suffer a capacity gap, due to the decommissioning of a nuclear power plant (Ignalina) and the maintenance of another one (Narva), in the face of rising demand.71 The authorities further took the view that the market would not provide the required capacity to fill the gap because of a number of reasons which made the building of such capacity uneconomical. Moreover, the existing interconnection capacity could not eliminate the problem, as the grid was unfit to bear the large volumes of electricity needed. Against this background, the Latvian authorities notified the Commission that they intended to grant aid by way of tender to an undertaking for the construction and

67

68 November 2013 Communication (n 34). Generation Adequacy SWD (n 20). 2009 Electricity Directive (n 3). 70 See Case 675/2009 Tender for Aid for New Electricity Generation Capacity (LV) [2010] OJ C/213 (the 2010 Latvian case). See also section 9.3.3 for a discussion on compensating the PSOs. 71 Laura Parmigiani, ‘Capacity mechanisms: EU or National Issue?’, Actuelles de l’Ifri, p 3: ‘Experience shows that early decommissioning and lack of visibility induced by changing legislation and world scale events (e.g. earthquake in Japan, shale gas in the US, CO2 market in Europe) might endanger the ability to satisfy the peak demand’. 69

20

EU Policy on Capacity Mechanisms

operation of new base-load capacity in the form of a solid fuel/LNG thermal power plant. In its 2010 state aid decision, the Commission agreed with the Latvian authorities and did not raise any objections. The winner of the tender would be obliged to operate the plant at least for 6,000 hours per year. In return, the winner would receive capacity payments between 2015 and 2025. These payments would, in essence, cover the fixed costs of the plant, and the tender mechanism would ensure that these costs were kept to a minimum. In any event, the Latvian authorities had envisaged a clause to terminate the payments earlier, should the price in the energy-only market raise above a certain threshold for a consecutive period of two years. In contrast with the Irish case, however, the Commission did not consider the Latvian mechanism to impose a PSO on generators and thus, did not analyse it according to the Altmark criteria. In the view of the Commission, the Latvian measure differs from the Irish measure because ‘[it] also encompasses support of normal capacity’, not just peak capacity.72 As a consequence, the Commission analysed the measure on the basis of Article 107(3) of the Treaty on the Functioning of the European Union (TFEU)73 and found the measure to be compatible aid.

1.4.3 The 2011 Estonian case Estonia is virtually the only country in the world using oil shale for wide-scale electricity generation. Although widely available in Estonia, oil shale is one of the most CO2-intensive fuels. In 2011 Estonia planned to grant state aid of up to €75 million annually over twenty years (ie a maximum of €1.5 billion) for the operation of two oil shale power plants in order to increase its locally-installed electricity generation capacity between 2013 and 2016.74 The beneficiary of the aid would have been the publicly-owned incumbent Eesti Energia, which was actively involved along the whole value chain from mining to electricity distribution, with shares of more than 75 per cent in all of these activities in Estonia. When opening the formal investigation procedure based on Article 108(2) TFEU, the Commission acknowledged that the measure could be justified on security of supply grounds. However, the Commission harboured serious doubts as to the proportionality of the measure. In particular, as Estonia did not plan to use a tender for the selection of the operator, there was a significant risk of overcompensating Eesti Energia. Moreover, the aid might have crowded out existing plants and discouraged investments in new plants both in Estonia and in neighbouring markets, since competing plants, contrary to the subsidized new plants, would be obliged to bear their full investment costs. In the end, Estonia withdrew the measure and the Commission closed the proceedings.

72 The 2010 Latvian case (n 70) at note 9. Unfortunately, the decision does not discuss this point any further. 73 Treaty on the Functioning of the European Union [2012] OJ C 326/47 (hereinafter ‘TFEU’ or ‘the Treaty’). 74 Case SA.30531, Aid for Capacity Payments for Oil-Shale Fuelled Electricity Production (ET) [2011] OJ C/235. As of 1 February 2015, the public version of the decision was not yet available due to confidentiality reasons.

1.4 EU approach after 20/20/20 Package

21

1.4.4 The 2009 RES Directive and its impact on RES development The 20/20/20 Package also included the new RES Directive, which replaced the 2001 RES Directive75 and imposed on Member States mandatory national RES targets.76 In response, Member States implemented a range of support schemes for RES capacity. According to a recent study by the Council of the European Energy Regulators (CEER) covering seventeen countries, this support amounted to €25.2 billion in 2010.77 Fuelled by state subsidies, RES-installed capacity across the EU increased exponentially. However, as noted earlier, the increasing share of RES capacity in the system not only exacerbates existing ‘missing money’ problems but also creates a new market failure, as flexible back-up capacities have no incentives to remain in the market. Against this background, in 2012 the Commission opened a public consultation on generation adequacy.78

1.4.5 The 2012 consultation on generation adequacy In its Consultation Paper opening the public consultation on generation adequacy (the 2012 Consultation Paper),79 the Commission acknowledged the significant impact of RES capacity on energy-only markets and invited interested parties to provide their feedback on possible solutions, including, notably, capacity mechanisms. The Commission collected 148 responses from Member States, industry as well as interested organizations and citizens.80 In general, respondents were uncertain as to whether energy-only markets can deliver the required investments to ensure generation adequacy and security of supply.81 However, there was unanimous recognition that the low marginal pricing of RES had added to the traditional missing money problem by undermining the business case for conventional generation.82 There was also broad consensus among stakeholders on the need for capacity mechanisms in the short to medium term, or at least until interconnection is built or the energy-only markets are 75 Directive 2001/77/EC of the European Parliament and of the Council of 27 September 2001 on the promotion of electricity from renewable energy sources in the internal electricity market [2001] OJ L 283/33 (the 2001 RES Directive). 76 Article 3 of 2009 RES Directive (n 66). 77 CEER, Status review of renewable and energy efficiency support schemes in Europe (Report C12-SDE33-03, 25 June 2013) pp 19–20. 78 http://ec.europa.eu/energy/en/consultations/consultation-generation-adequacy-capacity-mechanismsand-internal-market-electricity, accessed 1 February 2015. 79 European Commission, Consultation Paper on generation adequacy, capacity mechanisms and the internal market in electricity (the 2012 Consultation Paper), available at (n 78). 80 See Results of the consultation, available at http://ec.europa.eu/energy/en/consultations/consultationgeneration-adequacy-capacity-mechanisms-and-internal-market-electricity, accessed 1 February 2015. 81 Opinions differ even among the reports from within the same Member States: see the sharply contrasting replies of The Federation of Finnish Technology Industries (response of 29 January 2013, p 1) stipulating that current market prices give correct signals for investments in new generation capacity and that there is no proof of missing money in energy-only markets, and Finland (response of 15 November 2012, p 1) conceding that investors are reluctant to make investments with the current market prices even though there is a lack of capacity in some areas and that the market price is in many cases distorted by state interventions such as price regulation and subsidized production. 82 See Results (n 80) p 4.

22

EU Policy on Capacity Mechanisms

able to provide sufficient signals for investment otherwise.83 At the same time, there was little interest in an EU-wide blueprint for capacity mechanisms. The Netherlands and France argued, for example, that an EU-wide capacity scheme would be premature given the state of the market,84 whereas the UK underscored the different needs of different electricity markets.85 It is also worth noting that the Commission’s approach presented in its 2012 Consultation Paper remains consistent with its assessment of capacity mechanisms under state aid rules. Namely, the Commission’s draft proposal for the new Energy and Environmental Aid Guidelines (draft EEAG 2014–2020) from 18 December 201386 includes a separate section on state aid for generation adequacy87 and some of the conditions for aid compatibility listed therein are almost identical to the points made in the Consultation Paper. For instance, point 1(a) of the Consultation Paper requires that the capacity mechanisms be clearly established in the context of the internal market, in particular increased interconnection. In the same line, para 209(d) of the draft EEAG from December 2013 requires Member States to assess the potential existence of interconnectors, and para 219 of the draft EEAG provides that the measure should not negatively impact the internal market. The new EEAG 2014–2020, eventually adopted on 9 April 2014,88 are further discussed in chapter 9 on state aid.

1.5 The November 2013 Communication According to the November 2013 Communication, capacity mechanisms are threatening market integration. They are thus tolerated, but only as an extrema ratio, ie once the potential of cross-border trade and demand side response has been exhausted and ‘other regulatory failures’, such as wholesale and retail price regulation, have been removed. This approach permeates the Commission’s ‘objective justification’ 83 For example, Dutch response, p 6, Czech response, p 5, French response, p 3, UK response, p 8, Finnish response, p 3, German response, pp 6–7 and other responses of public authorities, ENTSO-E response, p 14, CEER response, pp 12–14, Clingendael response, p 23, European Nuclear Energy Forum response, p 5 and other organizations. It is true, however, that some Member States are highly reluctant visà-vis the implementation at the Member State level of capacity mechanisms, especially in the fear of locking in old fossil fuels generators, see Swedish response, p 4, and Norwegian response, pp 2–3, for example. See also the International Energy Agency (IEA), according to which ‘[t]he practical considerations also matter, and implementing the measures of the basic package takes time— several years— a lead time not compatible with urgent security of supply concerns. In this respect, IEA countries face different situations. Most states have already developed or improved energy markets and will continue to do so. While some regions face urgent generation adequacy constraints (the UK and Japan, for instance), most of other countries have enough excess capacity, as a result of sluggish demand growth. In the latter case, reforming electricity market design is not seen as a pressing concern’, IEA response, p 5. 84 French response, p 6, Dutch response, p 8. 85 UK response, p 12, which noted that ‘[t]he needs of different electricity systems in the EU mean that designing a capacity mechanism blueprint that is suitable for all these different cases would be unlikely to succeed. Issues such as market structure, geographic barriers, level of interconnection, degree of demand side responsiveness and generation mix must be taken into account as capacity mechanisms are designed.’ 86 Paper of the Services of DG Competition containing draft Guidelines on environmental and energy aid for 2014–2020 (18 December 2013), available at http://ec.europa.eu/competition/consultations/2013_ state_aid_environment/draft_guidelines_en.pdf, accessed 1 February 2015 (draft EEAG 2014–2020, version of 18 December 2013). 87 Draft Guidelines on environmental and energy aid for 2014–2020 (n 86) p 60. 88 EEAG 2014–2020 (n 135).

1.5 The November 2013 Communication

23

discussion within the Generation Adequacy SWD: there is a seventeen-step Checklist entitled ‘justification of intervention’ covering the: (a) assessment of the generation gap; (b) causes of generation adequacy concerns; and (c) options other than support for capacity.89 The Commission’s misgivings towards capacity payments also influence the proportionality assessment,90 as illustrated by the extensive thirteen separate recommendations listed in the section ‘recommendations to avoid distortion of internal electricity market’.91 The Communication seems to impose new obligations on Member States to justify the need for state intervention by setting out multiple tests to be satisfied before implementing capacity mechanisms. This seems hardly consistent with the case law on Article 106 TFEU and appears to have an unduly restrictive impact on the possibility to invoke the Altmark case law. The Communication recognizes that ‘public intervention to promote generation adequacy may entail public service obligations imposed on generators, suppliers and/or TSOs’.92 It is well-settled that ‘in the absence of specific Union rules defining the scope for the existence of a service of general economic interest (SGEI),93 Member States have a wide margin of discretion in defining a given service as an SGEI and in granting compensation to the service provider.’94 In particular, the Commission can question a Member State’s choice only in case of manifest error of assessment.95 Thus, for instance, a number of companies challenged the Commission decision authorizing aid ex Article 106(2) TFEU96 inter alia because Spain had failed to prove that there was a genuine threat to security of supply. The General Court rejected these appeals, arguing that Spain had provided sufficient justification for a threat to security of supply

89

See Generation Adequacy SWD (n 20) pp 34–6. The discussion on avoiding distortions in the Generation Adequacy SWD (n 20) is framed in terms of least restrictive options, which is part of the proportionality assessment. See, for instance, Joined Cases C-204/12 to C-208/12 Essent Belgium NV v Vlaamse Reguleringsinstantie voor de Elektriciteits—en Gasmarkt (judgment of 11 September 2014, nyr) para 89, where the CJEU held that proportionality entails that a national measure ‘must be appropriate for ensuring attainment of the objective pursued and must not go beyond what is necessary in order to attain that objective.’ 91 See Generation Adequacy SWD (n 20) p 35. 92 See Generation Adequacy SWD (n 20) p 3 (emphasis added). 93 SGEI are commercial services of general economic utility, for example transport, energy and communications services, on which public authorities impose public service obligations (PSOs). See, http:// europa.eu/legislation_summaries/glossary/services_general_economic_interest_en.htm, accessed 1 February 2015. 94 See Communication from the Commission on the application of the European Union State aid rules to compensation granted for the provision of services of general economic interest [2012] OJ C 8/4, para 46. See also, for instance, Case C-265/08 Federutility and others v Autorità per l’energia elettrica e il gas [2010] ECR I-3377, para 29. 95 Federutility (n 94); see also Case T-289/03 BUPA and others v Commission [2008] ECR II-81, paras 166, 169, and 172 and the case law cited therein; Case T-17/02 Fred Olsen [2005] ECR II-2031, para 216, see also Communication from the Commission (n 94) para 46. 96 Article 106(2) of the Treaty provides the legal basis for assessing the compatibility of state aid for SGEIs. It states that undertakings entrusted with the operation of SGEIs or having the character of a revenue-producing monopoly are subject to the rules contained in the Treaty, in particular to the rules on competition. However, Article 106(2) of the Treaty provides for an exception from the rules contained in the Treaty insofar as the application of the competition rules would obstruct, in law or in fact, the performance of the tasks assigned. This exception only applies where the development of trade is not affected to such an extent as would be contrary to the interests of the Union. 90

24

EU Policy on Capacity Mechanisms

and that this justification was not manifestly incorrect. Hence, according to the General Court, the Commission was right to uphold Spain’s claim that the Spanish measure at issue was an SGEI.97 Against this background, the Communication’s insistence on the need for a Member State to pass a multitude of tests before concluding that a given capacity mechanism is necessary appears to be at odds with the case law on Article 106(2) TFEU. It is undisputable that capacity mechanisms address issues related to the security of electricity supply (in the form of adequacy between supply and demand, and the existence of suitable back-up capacity for RES). In this respect, and according to settled case law, security of supply is recognized as a valid public policy ground justifying state intervention.98 It follows that the Member State wishing to implement a capacity mechanism needs only to prove that there is a threat to security of supply because the market fails to deliver the necessary incentive for building new capacity and/or enough backup capacity to remain in the market. The Commission’s role in this context is limited: it can only check whether the Member States’ reasons are not manifestly wrong. As a consequence, the inability to meet one or more of the seventeen tests in the Generation Adequacy SWD should not necessarily constitute a legal obstacle preventing a Member State from enacting a capacity mechanism as an SGEI. These considerations are also relevant in the context of the Altmark case law. As the previously mentioned 2003 Irish CADA case shows, the first Altmark condition, according to which the recipient undertaking is required to discharge public service obligations and those obligations have been clearly defined, is tightly linked to the existence of a genuine SGEI. By restricting Member States’ ability to designate capacity mechanisms as SGEI, the November 2013 Communication seems also to be taking an unduly restrictive approach to the possibility for Member States to invoke the application of the Altmark case law. Finally, the arguments on the unduly restrictive nature of the seventeen-step approach are potentially relevant in the context of infringement proceedings ex Directive 2009/72. By its very wording, Article 3(2) of the Directive must be read together with Article 106(2) TFEU (‘Having full regard to the relevant provisions of the Treaty, in particular Article 86 thereof . . . ’). As a consequence, if the Commission used the seventeen-step approach to question the objective justification of capacity mechanisms of an SGEI in a given Member State in the context of an action based on infringement of the 2009 Electricity Directive,99 this could also be questionable.

1.5.1 Some of the recommended steps are questionable in and of themselves The Generation Adequacy SWD proposes that ‘Member States’ generation adequacy assessments need to take account of existing and forecast interconnector capacity as 97 See Order of the President of the General Court of 17 February 2011 in Case T-484/10 R Gas Natural Fenosa SDG v Commission (nyr). 98 See, ex multis Case C-242/10 Enel Produzione [2011] (nyr) paras 50–4. 99 2009 Electricity Directive (n 3).

1.5 The November 2013 Communication

25

well as the generation adequacy situation in neighbouring Member States. Surplus generation in neighbouring Member States may alleviate adequacy concerns.’100 In this context, the Generation Adequacy SWD recalls that ‘national generation adequacy assessments should be complemented by regional and Union-wide assessments’,101 recommending the use of the European Network of Transmission System Operators for Electricity’s (ENTSO-E)102 Scenario Outlook and Adequacy Forecast (ENTSO-E’s Report).103 The Generation Adequacy SWD makes also a related point: ‘Currently, there is overcapacity in many markets [ . . . ]. When faced with a structural generation overcapacity in the market, Member States may consider other measures [rather than setting up capacity mechanisms] such as facilitating exports by adding interconnection capacity or allowing for the retirement of environmentally inefficient plants, for example through the implementation of environmental legislation or by removing subsidies.’104 The Commission’s insistence on exploring the potential of the internal market before implementing capacity mechanisms calls for a number of comments. First, as the Generation Adequacy SWD itself notes, the ENTSO-E report suffers from methodological shortcomings and cannot be used as a basis for a reliable adequacy assessment because TSOs apply different methods for the calculation of the margin peakload and RES are treated differently.105 Second, the existence of interconnection infrastructure is a necessary, but not a sufficient condition for electricity flows to take place. What counts is the ‘deliverability’ of capacity across electricity systems. This requires TSO cooperation: only when a TSO in country A is able to direct a generator in country B to make its electricity available when needed, can interconnection be a real substitute to capacity mechanisms. However, these forms of cooperation are usually not in place, and, when they do exist, are voluntary and limited in scope.106 Third, besides the different ability of Member States to rely on interconnection due to their geography, electricity flows between countries depend on price differences.107 However, the existence of such price differences and the related flows do not necessarily go in the direction of solving generation adequacy. In fact, sometimes such flows may be the cause of generation adequacy problems. For instance, in the previously mentioned

100

101 Generation Adequacy SWD (n 20) p 6. Generation Adequacy SWD (n 20) p 7. ENTSO-E stands for the European Network of Transmission System Operators for Electricity, an association of Europe’s transmission system operators (TSOs) for electricity. 103 See the most recent ENTSO-E’s Report, Scenario Outlook and Adequacy Forecast 2014–2030 (June 2014). 104 Generation Adequacy SWD (n 20) p 9. 105 Generation Adequacy SWD (n 20) p 7 and note 18. 106 The Communication welcomes ‘the joint declaration that the Member States of the Pentalateral Forum recently issued’ (Generation Adequacy SWD (n 20) p 7). However, this declaration explicitly relies on voluntary cooperation between TSOs. Moreover, even this form of cooperation falls short of what would be needed for a TSO in country A to be able to control a generator in country B. See the Political Declaration of the Pentalateral Energy Forum of 7 June 2013, which mainly provides for cooperation and coordination on the basis of exchange of information and explicitly excludes, in Annex 3, any legal effects of the Declaration under international law. 107 Phil Baker and Meg Gottstein, ‘Capacity markets and European market coupling—can they coexist?’, Regulatory Assistance Project (RAP) Discussion Draft of 13 March 2013, p 10. 102

26

EU Policy on Capacity Mechanisms

2010 Latvian case, the Commission recognized that there were cheap imports from Belorussia flowing into Latvia.108 Because of these cheap imports, investors did not have an incentive to build new capacity. However, the interconnection lines were not suited for sustaining the delivery of large volumes of electricity on a continuous basis for an extended period of time. Thus, the existence of cheap imports across the border created a ‘missing money’ problem, which in turn justified a tender for new capacity. The same can be true in the mirror case of an exporting country. If generators in country A tend to export their electricity to country B where prices are higher, country A may suffer from a capacity issue, for which it cannot count on local generators. Thus, the price differential and the existence of exports to country B may in fact trigger the need for country A to take action. In the context of the state aid decision on the UK capacity mechanism,109 a report commissioned by OFGEM noted that ‘interconnector flows have helped to reduce the number of GB (Great Britain) low capacity margin hours in a year. However, for the hours of highest GB system stress (ie where capacity margins are below 10 per cent) interconnection flows have not consistently helped and have sometimes worsened capacity margins in GB.’110 Price differences, moreover, can be artificial, and thus distort trade flows. A combination of low demand and increased deployment of RES has pushed wholesale prices down in some Member States, including Germany, Belgium, and Spain.111 As a result, these countries could potentially seek to export electricity to neighbouring States where prices are higher. However, this does not alleviate their capacity concerns. As a matter of fact, Spain has already implemented capacity payments (to remunerate investment in capacity and for back-up capacity)112 and Germany may soon follow suit.113 As to the importing countries, it is unclear the extent to which they would be well advised to rely, in order to address potential security of supply concerns, on import flows which stem from dysfunctions in the exporting Member States’ wholesale markets. Finally, the time-scale for building interconnection may not be compatible with security of supply concerns. It is well known that building generation facilities is faster than building interconnection.114 Cognizant of these difficulties, the Commission has pushed for a new regulation on trans-European energy infrastructure.115 However, this Regulation entered into force in June 2013 and its period of implementation has only just begun. Thus, the benefits of the project under the aegis of this Regulation may not be realized before 2017–2018. This time frame would be hardly compatible with the short to medium-term capacity needs in some Member States.

108

109 The 2010 Latvian case (n 70) para 6. UK capacity mechanism (n 49). UK capacity mechanism (n 49) para 119 (emphasis added). 111 112 See Generation Adequacy SWD (n 20) p 9. See chapter 21. 113 See chapter 15. 114 Laura Parmigiani (n 71) p 3. According to the Generation Adequacy SWD (n 20) p 27, it generally takes four years to build a new generation plant. By contrast, by way of example, the expansion of an interconnection facility between France and Spain was planned in 2008 and its construction has still not been completed (1 February 2015). 115 Regulation No 347/2013 on guidelines for trans-European energy infrastructure [2013] L 115/39 (Regulation on trans-European energy infrastructure). 110

1.5 The November 2013 Communication

27

1.5.2 The emphasis on demand side response The Commission ascribes to demand side response the ability to deliver around 60 GW of capacity, or the equivalent of sixty nuclear power reactors. However, other than its sheer costs and practical difficulties of implementation,116 demand side response may raise state aid and competition policy concerns. The opinion of the French competition authority on the French capacity mechanism from December 2013 clearly illustrates this problem.117 The new French law on demand side response118 envisages the creation of a new product to be traded mainly in the wholesale balancing market: blocks of off-switchable capacity. In particular, the TSO could sell blocks of off-switchable capacity at the wholesale balancing market after having identified demand side response opportunities by their clients. Thus, the TSO could procure reserve margin from: (a) generators and (b) demand side operators offering the new product. The demand side operators would, in turn, have to identify willing clients able to offer their readiness to be disconnected which is suitable for trading in the balancing market (eg disconnection at short notice with no risk for the safety of the willing client, etc). The demand side operators would also have to negotiate the terms and conditions of disconnection with their registered customers, including the remuneration of the willing clients. Besides revenues from the sale of the product on the balancing market, the demand side operators would also receive a premium from the state deposit and consignment office (Caisse de Depôts et Consignations). The premium is funded through a levy imposed on all customers. According to the French competition authority, the premium may constitute state aid for the demand side operators.119 Moreover, the French competition authority has also expressed concerns about the potential anti-competitive effects of the proposed regulation. In particular, it was said that EDF might leverage its market power in the market for retail supply into the demand response market through tying, that is, offering to its existing customers supply services together with a demand side response service. Finally, the French competition authority expressed doubts as to the environmental effect of the measures. A consumer switching off a heating appliance

116 In the UK capacity market decision (n 49) para 131, the Commission admits that demand side response is still in its ‘infancy’. 117 Autorité de la concurrence, Avis No 13-A-25 du 20 décembre 2013 concernant l’effacement de consommation dans le secteur de l’électricité, available at http://www.autoritedelaconcurrence.fr/pdf/avis/ 13a25.pdf, accessed 1 February 2015. 118 Law No 2013-312 of 15 April 2013 (Loi n 2013-312 du 15 avril 2013 visant à préparer la transition vers un système énergétique sobre et portant diverses dispositions sur la tarification de l’eau et sur les éoliennes) French Official Journal, 16 April 2013. 119 Avis No 13-A-25 (n 117) para 85: ‘Au surplus, le dispositif de prime semble à première vue susceptible de constituer une aide d’État au sens de l’article 107 (1) TFUE qui exige, pour retenir une telle qualification, que quatre conditions cumulatives soient remplies: (a) la mesure doit être financée directement ou indirectement par des ressources d’Etat; (b) la mesure doit être sélective; (c) elle doit conférer un avantage; (d) la mesure doit affecter ou menacer d’affecter la concurrence et les échanges entre États membres.’ See also para 82, where the authority questions the need for such premium when: (a) EDF already offers the same service without receiving any such premium; and (b) in Switzerland, firms offer a similar service (household power management) on a commercial basis.

28

EU Policy on Capacity Mechanisms

using electricity might switch on another using fossil fuel. The environmental benefit would thus be lost.120

1.5.3 Removing regulatory failures The Generation Adequacy SWD warns that ‘[b]efore deciding on public intervention to support generation adequacy, the causes of any investment gap should be objectively analysed. Where existing public intervention causes or exacerbates an investment gap, it may be more cost-efficient to review and adjust them.’121 Among the chief regulatory failures that Member States should remove (before considering that capacity mechanisms are necessary), two feature prominently in the Generation Adequacy SWD— wholesale and retail price regulation. However, on its analysis of these two regulatory measures, the Generation Adequacy SWD may have overlooked the following. As to wholesale price regulation, even markets with no price caps can fail to deliver generation adequacy. This is because, due to the low elasticity of supply, as capacity becomes scarce, the electricity market does not always clear and in a blackout situation there is no market price. In short, energy-only markets with no wholesale price regulation may not optimize blackouts, as all available generators produce as much electricity as they can, yet—regardless of the price—not all demand can be served.122 Moreover, after the large-scale deployment of renewables, the problem is that prices may be too low to reflect generators’ real operating costs. Thus, even if wholesale price regulation were to be removed, it would hardly have any impact on the energy-only market ability to send the correct signals to attract new capacity and retain the existing capacity for back-up purposes. As to retail price regulation, leaving aside the issue of vulnerable consumers, Article 3(3) of the 2009 Electricity Directive requires ‘that all household customers, and, where Member States deem it appropriate, small enterprises enjoy universal service, that is the right to be supplied with electricity of a specified quality within their territory at reasonable, easily and clearly comparable, transparent, and non-discriminatory prices.’123 It is thus unclear to what extent Member States can remove retail price regulation if they cannot fulfil the Directive’s mandate to guarantee reasonable prices for all household customers and, in some cases, small enterprises.124 In fact, the Commission itself seems to have taken a less belligerent approach towards retail price regulation. For instance, the Commission opened an infringement procedure against Italy in 2006 for its regulation of electricity and gas retail prices, a practice which infringes EU law. Several other Member States were notified with similar charges.125 Italy explained that its intervention on retail prices was very limited. Namely, the Italian regulation instituted a central buyer (the Acquirente Unico) which bought electricity on the market (as any large customer would do), in order to resell to 120 122 123 124

121 Avis No 13-A-25 (n 117) p 16. Generation Adequacy SWD (n 20) p 12. Cramton, Ockenfels, and Stoft (n 5) p 28. Article 3(3) of the 2009 Electricity Directive (n 3). See also Stephen Tully, ‘The human right to access electricity’ (2006) The Electricity Journal 19(3),

30–9. 125

European Commission, Press release IP/06/430, 4 April 2006.

1.5 The November 2013 Communication

29

the legacy supplier at a price which was in essence the price paid on the wholesale market (plus a mark-up to cover the Acquirente Unico’s operating costs, ie the cost of the staff needed to carry out the purchase). The legacy suppliers, in turn, had a duty to supply those customers who did not choose a new supplier after liberalization. In that respect, as argued by the Italian government, the only ‘regulated’ element was the legacy suppliers’ obligation to set the price for final customers on the basis of the price paid to the Acquirente Unico. In 2012 the Commission closed the investigation procedure without any finding of infringement. The decision is not public, but it is reasonable to infer that, according to the Commission, Italian retail price regulation did not go beyond that which was necessary for the purposes of guaranteeing the provision of a universal service. In fact, in a 2012 report on transparency in EU retail energy markets,126 drafted by a working group convened by the Commission, Italy featured as one of the model countries.

1.5.4 Cross-border participation in capacity mechanisms Among the proportionality requirements identified by the Commission, the Generation Adequacy SWD lists a series of actions Member States should undertake to avoid ‘distortions of the internal electricity market.’ As such, the Generation Adequacy SWD puts forward that ‘[capacity] mechanisms should be open to any capacity, including capacity located in other Member States, which can effectively contribute to meeting the required generation adequacy standard and security of supply.’127 First of all, this point suggests that there should be cooperation between TSOs, as otherwise it would be impossible for cross-border capacity to ‘effectively contribute’ to solving generation adequacy problems. However, even the Commission recognized, in the context of the state aid decision on the UK capacity mechanism, ‘the complexities of effectively allowing cross-border participation in a capacity mechanism’.128 Second, the effectiveness of the cross-border participation is subject, first and foremost, to the availability of interconnection capacity. One could imagine that a generator participating in a cross-border capacity mechanism should be granted priority of access to the available interconnection capacity, precisely to ‘effectively contribute’ to addressing the generation adequacy problem. However, it is difficult to see how such priority access could be justified under the case law in Vereniging voor Energie and other.129

126

Working Group Report on Transparency in EU Retail Energy Markets, 5th Citizen’s Energy Forum, November 2012. 127 Generation Adequacy SWD (n 20) p 28. 128 UK capacity market (n 49) para 136. The Commission accepted that foreign capacity could be excluded in the first capacity auction, albeit welcoming the UK commitment to facilitate ‘the participation of interconnectors from 2015’. 129 Case C-17/03 Vereniging voor Energie, Milieu en Water and Others v Directeur van de Dienst uitvoering en toezicht energie [2005] ECR I-04983. See also the Commission’s interpretative note on this case, Commission Staff Working Document on the decision C-17/03 of 7 June 2005 of the Court of Justice of the European Communities, Preferential Access to Transport Networks under the Electricity and Gas Internal Market Directives Brussels (SEC(2006) 547, 26 April 2006).

30

EU Policy on Capacity Mechanisms

In this case, the Dutch TSO reserved, on a preferential basis, a portion of the interconnection capacity for the Samenwerkende Elektriciteits Productiebedrijven NV (SEP), which, prior to liberalization, had an import monopoly and was entrusted with services of general interest. According to the Dutch authorities, security of supply justified the reservation of capacity, as the imports were linked to long-term contracts concluded by the SEP before market liberalization in fulfilment of its public mandate. However, the Court of Justice of the European Union (hereafter: CJEU, Court) dismissed this justification. The Court held that the Directive prohibited discrimination in the access to interconnection capacity. Any exception to this rule required the submission of an application for a transitional regime, but the Dutch authorities had not requested such a derogation. These types of legal obstacles—besides those of a technical nature—cannot be overlooked when calling for cross-border availability of capacity mechanisms.

1.5.5 Technological neutrality and environmental protection In the Commission’s view, ‘in order to be cost-effective and in line with Union policy on the environment, capacity mechanism should be open to all technologies able to solve the problem of an identified gap in generation adequacy.’130 At the same time, the Generation Adequacy SWD is adamant that ‘removing environmentally [ . . . ] harmful subsidies, including for fossil fuels may help correct market signals and reduce the need for further intervention.’131 Read together, these statements should provide comfort to Member States seeking to enact capacity mechanisms which exclude highly polluting technologies. Indeed, it would be paradoxical if capacity mechanisms could be used to keep afloat obsolete and environmentally harmful plants. In this respect, the Generation Adequacy SWD call to ‘assess holistically’ capacity mechanisms, in particular with respect to their impact on environmental target, is welcome.132

1.6 Conclusion—where next with capacity mechanisms in the EU? Prior to the November 2013 Communication, the EU approach to capacity payments focused on the ‘missing money problem’ and capacity tenders to attract investments in new capacity. The November 2013 Communication and the Generation Adequacy SWD reflect the impact of vast RES deployment and the emergence of a new market failure: the need to have available back-up capacity for the safe integration of RES. Capacity payments thus take on board the fundamental task of bridging the EU environmental ambitions with security of supply. Given the key role that capacity mechanism can play in this regard, it is regrettable that the Commission has thus far failed to adopt clearer ex ante guidance. The 130 131 132

Generation Adequacy SWD (n 20) p 26. Generation Adequacy SWD (n 20) p 14. Generation Adequacy SWD (n 20) p 17.

1.6 Conclusion—where next in the EU?

31

Generation Adequacy SWD states that the Commission services are ‘fully committed to working with Member States and regions through the Electricity Coordination Group and bilaterally with a view to effectively and efficiently addressing generation adequacy concerns.’133 However, it is unclear to what extent cooperation through the Electricity Coordination Group (or bilateral talks) can give to Member States the level of legal comfort needed. In any case, infringement procedures or state aid investigations do not appear to be the desirable avenue to address these important issues. In the 2014 Communication on the 2020–2030 climate and energy policy,134 the Commission envisaged the adoption of the so-called National Energy Plans (NEPs). An NEP is a holistic instrument through which Member States inform the Commission (and other Member States) of their progress in the energy field, spanning its different facets (RES targets, energy efficiency, carbon reduction, etc). The NEPs are submitted by Member States individually, but subject to consultation among the neighbouring states. These regional consultations permit each Member State to consider its goals in an integrated fashion, taking into account the realistic possibilities of its energy region. The NEPs are then approved by the Commission. A similar solution should also be possible for Member States wanting to enact capacity mechanisms. Considering that capacity mechanisms are often needed on a short to medium-term basis, this would avoid the protracted uncertainty of state aid and/or infringement proceedings, as well as the inherently adversarial approach likely to be adopted in these procedures. However, for this solution to be workable, the Commission should dismiss the over-prescriptive approach of the Generation Adequacy SWD. First, Member States owe a duty to their citizens to guarantee security of supply. Second, each Member State has different potentials for interconnection, demand side response, and the removal of regulatory barriers. Finally, the internal market dimension of capacity mechanisms and the need to integrate the objectives of the internal market, competition and environmental protection can be better preserved through consultations on a regional basis, between the neighbouring Member States. In other words, an excessively formal and one-size-fits-all approach may be impracticable in the current state of electricity market integration.135

133

Generation Adequacy SWD (n 20) p 5. The Group is to provide a platform for strategic exchanges between Member States, national regulatory authorities (NRAs), ACER, ENTSOE, and the Commission on electricity policy. 134 Communication from the Commission to the European Parliament, the Council, the European Economic and Social Committee and the Committee of the Regions, A policy framework for climate and energy in the period from 2020 to 2030 (COM/2014/015 final, 22 January 2014). 135 On 9 April 2014 the Commission published the new Guidelines on State aid for environmental protection and energy 2014–2020, hereinafter referred to as ‘EEAG 2014–2020’ (Communication from the Commission, Guidelines on State aid for environmental protection and energy 2014–2020 [2014] OJ C 200/ 1–55). The final version does not substantially differ from the draft put to consultation in December 2012 (n 86). Thus, the EEAG 2014–2020 and the November 2013 Communication (n 34) build on the same approach. For a thorough discussion of the EEAG, see section 9.4.

2 The Regulators’ View ACER’s Report on Capacity Mechanisms and the EU Internal Electricity Market Alberto Pototschnig and Martin Godfried

2.1 Introduction The Third Energy Package1 has set rules for the integration of national markets into a single, internal market for electricity and gas. This should ensure, inter alia, an efficient use of existing generation capacity, demand-side resources and cross-border transmission infrastructure. The European Council, in February 2011,2 set the target of completing the internal energy market by 2014, a goal reaffirmed in May 2013.3 At the same time, there is a growing concern in several Member States that electricity markets with increasing shares of intermittent RES generation, will not be able to deliver sufficient capacity to meet future demand at all times.4 The political sensitivity to blackouts, as well as uncertainty5 over whether investors will build new generation capacity, has compelled a number of Member States to intervene by introducing capacity mechanisms in order to ensure that a sufficient amount of capacity will be available. A capacity mechanism aims to provide market participants with a more effective stimulus to invest than that delivered by energy-only markets. Namely, it offers investors a more certain and stable stream of revenues in the form of capacity remuneration.

1

Directive 2009/72/EC of the European Parliament and of the Council of 13 July 2009 concerning common rules for the internal market in electricity and repealing Directive 2003/54/EC [2009] OJ L 211/ 55 (2009 Electricity Directive), Directive 2009/73/EC of the European Parliament and of the Council of 13 July 2009 concerning common rules for the internal market in natural gas and repealing Directive 2003/55/EC [2009] L 211/94 (2009 Gas Directive) and Regulation (EC) No 714/2009 of the European Parliament and of the Council of 13 July 2009 on conditions for access to the network for cross-border exchanges in electricity and repealing Regulation (EC) No 1228/2003 [2009] OJ L 211/15 (2009 Cross-border Regulation). 2 European Council, Conclusions on Energy (PCE 026/11, 4 February 2011) in particular point 4, available at http://www.consilium.europa.eu/uedocs/cms_Data/docs/pressdata/en/ec/119141.pdf, accessed 1 February 2015. 3 European Council, Conclusions (EUCO 75/1/13 REV 1, 22 May 2013) in particular point 2, available at http://www.consilium.europa.eu/ueDocs/cms_Data/docs/pressData/en/ec/137197.pdf, accessed 1 February 2015. 4 This objective has to be interpreted in probabilistic terms, ie as a given probability of meeting demand at all times, including at peak times. 5 At the moment, there is considerable discussion, and different views, as to how generation adequacy should be addressed in the context of the internal market, taking into account the necessary transition to a low-carbon energy system. For a discussion on the EU approach, see also chapters 1 and 4.

2.2 Contribution of energy-only markets

33

However, to the extent that these revenues are higher than those deriving from an energy-only market, the mechanism may impose additional costs to energy consumers.6 Many national electricity wholesale markets are highly interconnected. Adjacent electricity systems in regional(ized) electricity markets interact physically and economically. Therefore, capacity mechanisms can potentially distort cross-border trade or act as a barrier to trade if they are designed without considering cross-border impacts or implemented at national level without coordination with neighbouring jurisdictions.7 In view of recent and future developments, ACER has looked at the impact of different capacity mechanisms on the functioning of the internal market. Following its Opinion on Capacity Markets, provided to the European Parliament’s Industry, Research and Energy (ITRE) Committee in February 2013 (ACER’s Opinion),8 in July 2013 ACER issued a Report on Capacity remuneration mechanisms and the internal market for electricity (ACER’s Report).9 As discussed in the previous chapter, the Commission has also intervened recently to provide guidance to Member States on how to design and implement capacity mechanisms, in order to ensure the supply of electricity without jeopardizing the benefits offered by the EU-wide energy market.10 In particular, the Commission provided a checklist which allows Member States to verify the efficiency of their intervention and to improve it where necessary.11 This chapter illustrates the main conclusions presented in ACER’s Report of July 2013. The reader is referred to the Report for a more complete treatment of the impact of capacity mechanisms on the internal market.

2.2 The contribution of energy-only markets to generation adequacy In a pure energy-only market, in theory and in the absence of market failures,12 the operating costs (eg fuel, start-up costs) and capital costs of a plant are recovered exclusively through market prices for electricity and for the associated ancillary services.

6 The way in which any cost of the capacity mechanisms are recovered depends on the specific form the mechanism takes, but, typically, the bill ends up being paid by energy consumers. 7 For a more detailed discussion see chapters 3, 5, and 6. 8 ACER, Opinion of the Agency for the Cooperation of Energy Regulators No 05/2013 of 15 February 2013 on capacity markets (ACER’s Opinion). 9 ACER, Report on capacity remuneration mechanisms and the internal market for electricity, 30 July 2013 (ACER’s Report). 10 See Communication from the Commission, Delivering the internal electricity market and making the most of public intervention, C(2013) 7243 final (November 2013 Communication) and Commission Staff Working Document, Generation Adequacy in the internal electricity market—guidance on public interventions, 5 November 2013 (Generation Adequacy SWD). 11 In particular, see section 1.5 on the 2013 November Communication (n 10) and the Generation Adequacy SWD (n 10). See also Communication from the Commission, Guidelines on State aid for environmental protection and energy 2014–2020 [2014] OJ C 200/1–55 (EEAG 2014–2020), section 9.4. 12 For instance, the absence of smart metering or, more generally, the absence of mechanisms/tools to develop DSR.

34

The Regulators’ View: ACER’s Report

In most hours of the year and under most circumstances, there is more available generating capacity than needed to meet demand. During these hours, assuming competitive market conditions, the energy market price, if allowed to vary unhindered, should reflect the marginal operating cost of the most expensive unit dispatched or the opportunity cost of any energy-limited hydro resources operating at the margin.13 In these hours, base-load and intermediate-load generators with operating costs lower than the market price can recover their variable operating costs and obtain an infra-marginal rent14 which can be used towards covering fixed costs. In some hours, however, the margin between available capacity and (peak) demand may tighten and electricity prices may rise above marginal operating costs to include a scarcity premium.15 During these (rare) occasions of capacity shortage, the system experiences extremely high prices, potentially up to VOLL.16 During these hours, all plants in the merit order17 receive a price which also contributes to the recovery of their fixed costs (‘peak-load pricing theory’, for a graphical illustration see Figure 4.1). In an energy-only market, scarcity prices may attract investment in new capacity and prevent existing capacity from leaving the market if they are sufficiently frequent and sufficiently high. In the absence of such price spikes and without any other revenues (eg from the provisions of ancillary services), existing peak plants may exit the market without being replaced. This would reduce the available generation capacity and increase the probability of scarcity conditions and the frequency of scarcity prices. But as long as demand is sufficiently price responsive, and falls to zero at VOLL,18 an energy-only market will always deliver an equilibrium. The interaction between available capacity and demand determines the economically optimal level of installed capacity through the prices established in the market. The level of adequacy19 is therefore determined endogenously by the market. The ‘political’ acceptability of the adequacy provided by energy-only markets depends on the frequency with which prices reach very high levels (possibly approaching VOLL), and the ‘political’ implications of such high prices. It is the ‘political’ unacceptability of extremely high prices in energy-only markets which pushes Member States to intervene, eg by introducing capacity mechanisms, in order to reduce the frequency and level of price spikes.

13

Energy-limited resources are, for example, hydro storage or certain pump storage plants. Their generation capability is not only determined by the capacity of the plant, but also by the availability of primary energy, ie the kinetic energy associated with the volume of water available in the reservoirs. 14 The difference between the market price and the operating (ie variable) cost of the plant, also known as ‘producer surplus’. For a graphical illustration, see Figure 4.1. 15 Also known as ‘scarcity rent’. See section 1.1 for a discussion. For a graphical illustration, see Figure 4.1. 16 See a definition of VOLL, chapter 1 at n 7. 17 See section 1.1 for a brief explanation. 18 Since, given the definition of VOLL (chapter 1, n 7) no consumer is willing to pay a price for energy higher than VOLL. However, VOLL may be difficult to measure. 19 In fact, if demand is sufficiently price responsive, an energy-only market will deliver full adequacy, but at the cost of some demand ‘voluntarily’ reducing consumption when prices reach the VOLL level. It is this way of achieving market equilibrium and adequacy which may not be political acceptable.

2.3 Impact of capacity mechanisms

35

2.3 Impact of capacity mechanisms: design and distortions Section 1.2.3 presented an overview of capacity mechanisms according to ACER’s taxonomy (Figure 1.2) and the current developments in the EU (Figure 1.3). The different types of capacity mechanisms interact differently with energy-only markets and ancillary services mechanisms (balancing markets). If not properly designed, they may negatively impact the energy market, preventing it from producing reliable and efficient price signals. Chapter 1 also indicates that Member States tend to pay little attention to the potential impact of their national measures on the EU internal market and crossborder trade. As a result, the opportunities presented by the internal market and crossborder trade for delivering adequacy are often overlooked. Consequently enduring distortionary effects can occur. Few studies focus on the cross-border distortions20 which can result from a unilateral introduction of a capacity mechanism in one Member State. Detecting these possible distortions is sometimes not straightforward, as the impact of capacity mechanisms may be interlinked with that of other (national) market design features. Moreover, most of the existing capacity mechanisms have been in place for only a relatively short period of time, therefore time series of data are not long enough to allow firm conclusions on medium and long-term effects to be drawn. However, these difficulties do not prevent an initial assessment of the way in which capacity mechanisms may create distortions in the internal market, as set out next. First, capacity mechanisms can impact prices in the short term, thereby altering production decisions (operation of power generating plants), cross-border flows and competition. When a national capacity mechanism does not fully consider nondomestic generation capacity, it may also impact cross-border competition at a wholesale level, introducing short-term price distortions. For instance, the threshold price in strategic reserves or capacity obligations21 acts effectively as a cap on the market price.22 Therefore, if it is set too low, it will prevent market prices rising to signal scarcity. In addition, the dampened price may be ‘exported’ to jurisdictions with energy-only markets, thus leading to the average income of generators in those markets being reduced as well. Another example of a possible distortion is when generators in a capacity mechanism receive payments which are determined in a way that affects their electricity generation bids into the market, while in a neighbouring energy-only market generators do not receive such payments. This may tilt the playing field for generators on either side of a border. Secondly, capacity mechanisms may influence investment decisions (investment in plants and their locations), with potential long-term impacts. Several simulation results 20 Two types of efficiency distortions are distinguished: static short-term and dynamic distortions. The former are related to whether the production of electricity is at least cost effective and whether prices reflect the cost of production, with given capacities, while the latter are related to the efficiency of new investments and their location. In addition to efficiency distortions there are also redistribution effects such as ‘spill-over effects.’ For a discussion on cross-border effects of individual capacity mechanisms see chapter 6. 21 In the case where the obligation is defined with respect to a threshold price. 22 For an overview of these two types see sections 1.2.3.1 and 1.2.3.3.

36

The Regulators’ View: ACER’s Report

in the literature23 show that after a capacity mechanism is implemented, it becomes the main driver (as opposed to energy prices) for investments in new electricity generation capacity. In countries without capacity mechanisms investments in generation decline and plants are decommissioned earlier. Furthermore, if a capacity mechanism excludes cross-border capacity, it may lead to overcapacity and over procurement of capacity in the country where the mechanism is operational with a detrimental impact on consumers. The benefits of and requirements for cross-border participation in capacity mechanisms are presented in section 2.4 below. Thirdly, capacity mechanisms may have welfare redistribution effects between interconnected markets. For instance, implementing a capacity mechanism to ensure generation adequacy in country A will likely benefit the generation adequacy of a neighbouring country, in particular if the mechanism induces much more generation capacity than would be efficient for providing the required level of adequacy. The cost of such overcapacity supported by the capacity mechanism is borne by the consumers of country A. In this way, consumers of country A end up paying for capacity contributing to generation adequacy in the neighbouring countries. The magnitude of this type of spill-over effect depends on the size of the capacity remunerated by the capacity mechanism and the degree of market integration. The more integrated the markets are—with overcapacities from capacity mechanisms—the higher the spill-over effect will be. More generally, as with all policies and measures, additional distortions may result from incorrect design or implementation of capacity mechanisms.24 Any capacity mechanism design requires that choices are made with respect to several characteristics.25 These choices may have a significant effect on the way in which capacity mechanisms impact energy markets, both in the short and the long term. The design of a capacity mechanism is influenced, among other things, by the methodologies applied for assessing generation adequacy and security of supply levels.26 Currently, these methodologies differ considerably between Member States, which hampers the comparability of their results. This poses an additional challenge when designing capacity mechanisms. In its Opinion,27 ACER stated that ‘[i]t is however essential that any such [capacity remuneration] arrangement does not unduly interfere or distort the functioning of the energy market and does not delay the completion of the [internal market]. In fact, it would be most desirable if any arrangement aimed at promoting adequacy or flexibility 23 Jeroen de Joode, Paul Koutstaal, and Özge Özdemir Ozdemir, ‘Financing investment in new electricity generation capacity in Northwest Europe’, ECN Policy Brief, ECN-0-13-022, May 2013. Johan Linnarsson, ‘Quantitative assessment for a European Capacity Market’, Fortum presentation (19 June 2012), available at http://www.elforsk.se/Documents/Market%20Design/seminars/CapacityMarkets/4_Fortum.pdf, accessed 1 February 2015. 24 See discussion in chapter 4. 25 The design of a capacity mechanism should take into account the existing market structure and its imperfections in order to avoid additional distortions to the functioning of the internal market. 26 Generation adequacy (criteria) refers to a long term targeted security of supply standard. This can be expressed in loss-of-load expectation (LOLE) with for example a targeted standard of three hours LOLE per year. Further, security of supply level refers to the amount of (short term) reserves available relative to the demand. See section 5.5.1 for an explanation of LOLE. 27 ACER’s Opinion (n 8) p 9.

2.4 Cross-border participation in capacity mechanisms

37

were to exert its effect only when and to the extent that energy markets cannot provide sufficient stimulus for the required investments, while having as little influence as possible on the energy markets at other times.’ In this respect, one critical design element of many capacity mechanisms (including strategic reserves, capacity obligations, and reliability options)28 which may have a significant (and possibly distortionary) impact on the energy market, is the level at which the threshold/strike price is set. In principle, the level of the threshold price (strike price) should discriminate between normal (albeit possibly tight) market conditions and situations of acute scarcity. Therefore, it is obvious that the threshold/strike price should be set below VOLL.29 However it is also essential that the threshold price is set well above any price level compatible with normal (albeit possibly tight) market conditions, and therefore (well) above the operating costs of the most expensive generating unit in the market. In fact, the threshold price should be set at a level which the market price would reach only in case of severe scarcity and when there are prospects of further increases towards VOLL. In the context of the considerations on the political acceptability of very high prices, the threshold price should be set at the highest politically acceptable price level.30 It is worth noting that such a level for the threshold price is typically much higher than the strike price of option contracts available in the market and used by market participants for risk-hedging purposes. Consequently, while being both option contracts by structure, reliability options and market option contracts serve different purposes and should not be confused.

2.4 Cross-border participation in capacity mechanisms As highlighted in ACER’s Opinion, ‘[i]n the case of national mechanisms, greater efficiency can be achieved and the distortion of the internal market minimised by allowing participation of adequacy and system flexibility resource providers located in other Member States, taking into account available cross-border capacity’,31 as promoted by the 2006 Security of Supply Directive.32 ‘This however requires’—so states the Opinion—‘that these resources will be allowed to contribute, directly or indirectly (through their TSOs) to adequacy and/or system flexibility in the Member States in whose mechanism they participate even at time of crisis also in the Member State in which they are located.’33

28

For an overview of capacity mechanism types, see section 1.2.3. A reference/strike price equal to or above VOLL would not serve much purpose, as in theory the market price should never exceed that level. In practice, the price may rise even higher, if consumers are not exposed to short-term prices and retailers have an obligation to serve their consumers. 30 See also discussion in section 7.6.2. 31 ACER’s Opinion (n 8) p 10. 32 Directive 2005/89/EC of the European Parliament and of the Council of 18 January 2006 concerning measures to safeguard security of electricity supply and infrastructure investment [2006] OJ L 33/22 (Security of Supply Directive). 33 ACER’s Opinion (n 8) p 10. 29

38

The Regulators’ View: ACER’s Report

Cross-border participation in capacity mechanisms does not necessarily require that cross-border capacity is set aside. However, it requires a strong coordination of national security of supply policies and the fulfilment of the following conditions. First, the TSO, in whose control area the capacity mechanism34 is implemented, must be able, directly or through the adjacent TSO, to monitor the actual availability of the (capacity) resources committed by the foreign capacity providers over the contracted period and ensure that these providers are able to offer the same level of commitment with respect to security of supply as the domestic providers. Secondly, efficient cross-border capacity allocation mechanisms must be implemented on all timeframes, in particular in the day-ahead, intra-day and balancing timeframes. Lastly, Member States have to accept that their domestic capacity providers are partly contracted to ensure the security of supply of a neighbouring country and guarantee that these providers will not be hindered in exporting at any moment in time. In particular during system stress, TSOs cannot deviate from their routine in offering cross-border capacity35 on both sides of the border.36

2.5 Conclusions and recommendations In an integrated European energy market, security of supply and efficient market functioning are no longer exclusively a national consideration, but should be addressed as a regional and pan-European issue. From this perspective, generation and, more widely, generation adequacy should be addressed and coordinated at regional and European level to maximize the benefit of the internal market and to avoid adverse distortionary effects. However, ACER observes in its Report that currently Member States have national, uncoordinated, and often diverging approaches to security of supply. The lack of crossborder coordination on the issue of security of supply has resulted in a patchwork of capacity mechanisms in Europe to the detriment of the market integration process. This might appear paradoxical given Member States’ efforts to complete the internal market through enhanced cooperation. The analysis of current national generation adequacy policies to date does not provide evidence of distortions specifically related to capacity mechanisms, as it is difficult to disentangle them from other market design inconsistencies. Nonetheless, capacity mechanisms may result in short- and long-term malfunctioning of the internal market. In the short term, capacity mechanism design may affect the natural price formation in the energy market (eg the bids for energy). In the long term, they may lead to excessive procurement of capacity, if cross-border capacity is not 34 Reliability options may be an exception to this, since this mechanism provides incentives to capacity availability at time of scarcity through the prospect of the financial contractual payments, typically linked to the difference between the market clearing price and the option strike price, which contracted resources may be called to make. Therefore, in the case of reliability options, TSOs do not need to control the availability of contracted resources. 35 Without such a guarantee, a foreign capacity provider would not be able to deliver the same level of commitment with respect to security of supply as a local provider. 36 In particular, Member States have to comply with Article 4(3) of the Security of Supply Directive (n 32). For an in-depth discussion on cross-border effects of capacity mechanisms, see chapter 6.

2.5 Conclusions and recommendations

39

appropriately considered. This would have a directly detrimental impact on customers through higher energy retail prices. Given that capacity mechanisms might not be easily reversible, their implementation requires a careful ex-ante assessment of various aspects, including cross-border flows, competition, prices, and investments. In addition, the implementation of capacity mechanisms should not delay the completion of the internal market: the removal of barriers to the well-functioning of energy markets and to the formation of reliable and efficient price signals across Europe should remain a priority. ACER believes that the risk of potential distortions can be addressed by better coordinating security of supply measures. Further, where capacity mechanisms are considered, the potential cross-border distortions should be assessed based on a common set of criteria. Consequently ACER proposed the following recommendations. First and foremost, generation adequacy criteria and security of supply levels should be harmonized where possible. This would create common definitions underlying security of supply, and should go hand in hand with the implementation of a common (at least regional) and coordinated approach to security of supply assessment.37 Secondly, national capacity mechanisms should assure participation—to the extent possible—of adequacy and system flexibility resources provided by generators and load located abroad. Foreign participation would ensure greater efficiency and minimize distortion of the internal market. However, as explained in section 2.4 above and further discussed in chapters 4, 5, and 6, the challenges are significant in this respect. Further, harmonized or regional capacity mechanisms may reduce the need for foreign participation in national capacity mechanisms. Finally, where capacity mechanisms are introduced at a national level, their design should incorporate the most effective and efficient solutions without distorting the functioning of the internal market. To this end, and to improve the transparency of capacity mechanisms across the Member States, the introduction of national mechanisms should be accompanied by a sound and detailed impact assessment. This should take an internal market perspective and investigate the nature of the problem which the mechanism intends to address, as well as the necessity and appropriateness of the proposed mechanism. Moreover, it should look into whether and how cross-border capacity is taken into account and whether there are any possible short-term and long-term distortions of the internal market, including cross-border flows, competition, prices, and investments, and how they can be avoided or limited. The impact assessment on such measures should include the cost of the mechanism (including implementation costs) and the (estimated) costs of capacity remuneration.

37 Such a common approach would, for example, determine which generation resources are taken into account to deliver security of supply.

3 Capacity Mechanisms in the European Market Now, but How? Arthur Henriot and Jean-Michel Glachant

3.1 Introduction The ability of energy-only markets to provide generators with a revenue stream covering their fixed costs has recently been questioned in Europe. The debate on energy-only markets versus capacity mechanisms, or, more specifically, on the comparative advantages of the different capacity mechanisms,1 has already been the object of extensive research,2 but this old argument is now taking place in a new context. We identify two specific issues requiring more investigation. First, a large share of the resources remunerated will have to operate in a flexible way, so as to cope with the variability of intermittent RES. In this context, generation adequacy (defined as the availability of sufficient capacity resources when needed— including activation of demand response) is not only about securing a minimum reserve margin, but also about delivering an adequate flexibility mix for the system. Part of the incentives to promote generation flexibility (the ability to adapt production or consumption to the system needs within a given time frame) might be embodied in short-term energy prices, but it is clear that the issue of generation adequacy cannot be completely separated from the issue of flexibility. Second, the implementation of national capacity mechanisms in Europe may hamper the progressive convergence towards an integrated European electricity market.3 Different needs, different resources, and different priorities have led to a patchwork of capacity mechanisms, and it seems unlikely that a European mechanism could be devised which meets the needs of all Member States. Yet, it is vital that national schemes do not lead to self-sufficiency of Member States, ie the ability of Member States to ensure generation adequacy at all times without relying on electricity imports from their neighbours. Self-sufficiency is costly, as the benefits of sharing resources at a regional scale are lost. It will be all the more costly as it is impossible, in practice, to confine the impact of capacity mechanisms to generation adequacy: energy exchanges may also be distorted, which would lead to further inefficiencies. In order to mitigate the impact of national capacity mechanisms, it is important that these schemes remain open to contributions from cross-border resources. 1

For a taxonomy of capacity mechanisms, see above, section 1.2.3. A recent review of these contributions can be found in Paul L. Joskow’s symposium on ‘Capacity markets’ included in Economics of Energy and Environmental Policy 2 (2). 3 Jean-Michel Glachant and Sophia Ruester, ‘The EU internal electricity market: done forever?’ (2014) Utilities Policy 31, 221–8. 2

3.2 Pervasive impact on remuneration of flexibility

41

The objective of this chapter is, therefore, to discuss the implementation of capacity mechanisms in the context of integrated electricity markets with a high share of intermittent RES. Section 3.2 focuses on the relationship between flexibility and generation adequacy, and explains why these two conjoined issues cannot be separated in a system characterized by a high level of intermittent RES. Section 3.3 lists the costs of a self-sufficient capacity mechanism, considering its pervasive impact on remunerating flexible generation resources. Section 3.4 sets out the requirements for coordinating national generation adequacy policies at EU level.

3.2 The pervasive impact of capacity mechanisms on the remuneration of flexibility 3.2.1 Ensuring generation requires flexible resources . . . In a system with a high share of intermittent RES, generation adequacy is not limited to the capacity margin. As the output of these resources is variable and not perfectly predictable, the system will need flexible back-up resources. In this context, flexibility is defined as the capability of a given resource to adapt production or consumption within a given time frame (regardless of potential variations in RES production). Therefore, flexibility can be considered as the ability to start up and quickly ramp up power units, to cycle frequently, and to operate at low minimum loads.4 Different power systems have different requirements in terms of flexibility. These requirements will vary with the generation mix, the load pattern, the nature of RES, and interconnections with neighbouring countries. The case of hourly changes in residual demand (defined as the demand to be covered by generation sources other than intermittent RES) for both GB and Germany is illustrated by Bertsch et al.5 Their simulations of a European power system, up to 2050, show that the maximum hourly change in residual load (up and down) will increase as RES constitute a higher share of the generation mix. The first and third quartiles of hourly changes in residual demand, as well as the maximum hourly variations, are represented in the boxplot, in Figure 3.1. These results will vary considerably across countries of different sizes, with different resources and generation mixes, as can be seen with GB and Germany. For comparable average hourly changes, the maximum hourly changes will be twice as large in Great Britain as in Germany which, in these simulations, is due to more diversified renewable resources in Germany. Hence, flexible generation resources are needed in power systems, but flexibility is costly. Increased usage leads to higher costs for generators, while maintenance

4 Generation units are inflexible if they face significant ‘ramp constraints’. Ramp constraints limit the capability of units to change production over short periods of time, and include operating ramp constraints, also known as ramp-up and ramp-down rate limits, start-up ramp constraint and shut-down ramp constraint. 5 Joachim Bertsch, Christian Growitsch, Stefan Lorenczik, and Stephan Nagl, ‘Flexibility in Europe’s power sector—an additional requirement or an automatic complement?’, Beiträge zur Jahrestagung des Vereins für Socialpolitik 2013: Wettbewerbspolitik und Regulierung in einer globalen Wirtschaftsordnung—Session: Market Theory and Applications, No E14-V2.

–40000 –20000

0

20000

40000

Capacity Mechanisms in the European Market

Hourly residual load changes

42

DE2011

DE2020

DE2050

GB2020

GB2050

Figure 3.1 Hourly changes in residual demand (MW): actual for 2011 and model simulations for 2020 and 2050 Source: Bertsch et al (n 5). DE—Germany, GB—Great Britain.

contracts must be renegotiated to allow for more flexibility.6 Plants will operate in a flexible way provided that they receive incentives to do so, for instance through energy prices. Note that another interesting result of the study by Bertsch et al is that the maximum hourly variation of the residual load will increase much more than the standard deviation from these hourly variations. In practical terms, it means that flexible resources will face low load factors and uncertain remuneration. The same arguments have thus far been raised to advocate the introduction of capacity mechanisms, which implies that capacity mechanisms will remunerate flexibility. It is widely acknowledged that flexible resources will be needed in the future to cope with the development of intermittent RES. However, the issue of remunerating generators for their flexible operation is sometimes presented as separate from the generation adequacy issue. It is argued that some long-term capacity mechanisms could be implemented to ensure generation adequacy, while new short-term price signals could ensure flexibility. In practice however, it is difficult to see how a pure generation adequacy policy could be implemented. It is evident that a properly designed generation adequacy policy should ensure the availability of capacity where and when needed. It is also clear that remuneration for availability conversely implies the higher opportunity costs of unavailability. In a system featuring a high share of RES capacity must be available, in particular, at times when RES output quickly drops to low levels. In order to be available when needed, inflexible units (ie having long start-up times and slow ramping rates) have to start generating earlier, when RES are still available and prices are low (or negative). For a thermal unit with relatively high variable costs, being available when needed may hence imply generating at a loss. Investment in more flexible units, which 6 José Ignacio Pérez-Arriaga and Carlos Batlle, ‘Impacts of intermittent renewables on electricity generation system operation’ (2012) Economics of Energy and Environmental Policy 1 (2), 3–17.

3.2 Pervasive impact on remuneration of flexibility

43

are able to start up and ramp up quickly at less cost, would avoid such losses. The incentives to invest in flexibility will then be determined by the (high) value of generating when required, at times when intermittent (and cheap) RES are not available and the corresponding prerequisite of generating when prices are low (or negative) as RES are still available. This means that any generation adequacy policy raising the opportunity costs of unavailability also incentivizes a more flexible operation of resources. One might argue that a penalty for unavailability should be reserved for a situation when flexibility is not needed, whereby inflexible generators would be exempted from the penalty. Yet, the relevance of such clauses is highly questionable in a system characterized by rapid and significant variability of the residual load, as they would then greatly weaken the guarantees of such a generation adequacy policy.

3.2.2 The impact of capacity mechanisms will hence not be limited to ‘capacity’ There are different ways to remunerate flexibility. It can be remunerated implicitly, through energy prices. The value of flexibility must then be reflected by sufficient fluctuations of energy prices (high when energy is needed and low when energy is not needed), as well as through arbitrage opportunities between day-ahead, intraday, and balancing markets. This approach is, for instance, defended by Bertsch et al,7 who shows that there is no need for explicit remuneration of flexibility to cover the costs related to ramping constraints.8 In their simulation, participants receive sufficient incentives to be flexible in order to benefit from high energy prices, when possible. Alternatively, flexibility can be remunerated explicitly through the definition of new flexibility products. California, for example, has considered introducing new flexible ramping products that aim at ensuring sufficient ramping capacity in real-time dispatch to manage the grid.9 Another way of explicitly rewarding flexibility is by setting technical requirements matching the flexibility needs. In Ireland for instance, the grid code imposes a maximum ramp rate for generation units, which cannot be less than 1.5 per cent of registered capacity per minute.10 In practice, remuneration is often through a combination of these two options, whereby flexibility is partially rewarded through energy prices, and implicitly incentivized through the definition of new flexibility products established by the TSO (within a capacity mechanism or not). However, in a competitive long-term equilibrium, the remuneration of flexibility must be just sufficient enough to cover the costs of providing this flexibility (including a satisfactory profit margin). If the need for flexibility is constant, additional remuneration for the same service through one form will be compensated by a lower remuneration through another form (for instance, a lower differential in energy prices). Hence, a consequence of flexibility remuneration through a generation adequacy policy is that 7 9 10

8 Bertsch et al (n 5). See n 4 above for the definition of ramp constraints. California ISO, Flexible ramping products (Revised Straw Proposal August 2014). Eirgrid, Grid code version 5.0 (October 2013).

44

Capacity Mechanisms in the European Market

it will automatically reduce the remuneration of flexibility in other markets, such as the energy market or the market for ancillary services. Flexibility is not a well-defined product and it can be offered through a wide range of resources, both in terms of technology and location. The optimal set of resources required to meet future flexibility needs is still not fully identified and we rely on markets to find the optimal technologies and strategies to deal with these constraints. The distribution of remuneration for flexibility across the different markets is not relevant, as long as each market remains accessible to all potential flexibility providers. However, in practice, when compared to energy markets, the range (in terms of technology and location) of resources allowed to bid in capacity mechanisms is limited. Some resources can be disregarded as such (for instance, intermittent RES, demand-side resources, cross-border resources). In other cases, resources may be excluded if they do not meet certain requirements for participation. For example, the time lapse between calls for capacity and delivery can be incompatible with the development of transmission assets that often require a lengthy process from conception to operation. The availability requirements can also be incompatible with demand-side resources or intermittent RES, if these resources are required to be available at any time and for long uninterrupted periods of time. At times, limitations on the lowest level of bids might prevent the participation of smaller generation units. Finally, the means of assessing the contribution to generation adequacy may further discriminate against certain resources. As the value of flexibility is reallocated, from the energy and ancillary services markets to a capacity mechanism bound to limited resources, the potential (technology and location) to offer flexibility is inevitably reduced. This impact will be greater if the share of flexibility remuneration from the capacity mechanism is larger, and the mechanism more exclusive. This could be especially costly if there is uncertainty regarding the best set of resources required to meet future flexibility needs.

3.3 The costs of self-sufficient capacity mechanisms 3.3.1 Designing purely national capacity mechanisms is expensive . . . The needs, the resources, and the objectives of Member States differ substantially. France has to deal with high winter-peak demand due to its reliance on electric heating.11 Germany wants to phase out nuclear plants and needs more transmission capacity to carry northern wind power to the south. The Spanish CCGT fleet is currently unprofitable as a result of overcapacity.12 Such specificities have resulted in a wave of heterogeneous proposals for capacity mechanisms, from centralized, targeted strategic reserves in Belgium,13 to decentralized, market-wide capacity obligations on suppliers in France.14 Yet, the difficulty in designing a unique capacity mechanism suitable for all Member States is certainly not a reason to dismiss any consistency 11 Réseau de Transport d’Électricité (RTE), French capacity market. Report accompanying the draft rules (Mécanisme de capacité: proposition de règles et dispositions complémentaires), 9 April 2014, pp 26–7. 12 For a discussion on the Spanish electricity market see chapter 21. 13 14 See chapter 13. See chapter 14.

3.3 Costs of self-sufficient capacity mechanisms

45

across national capacity mechanisms. The benefits of a multilateral (or regional) approach towards security of supply are, indeed, obvious by comparison to a national self-sufficient approach.

3.3.1.1 Self-sufficiency leads to excess capacity First, the current level of reserve margin and future needs vary across Member States. Oftentimes, countries with a large amount of surplus capacity are neighbouring countries with diminishing reserve margins. This problem has recently been studied by the THEMA consulting group on behalf of the Commission’s Directorate-General Energy. According to their results, presented in Figure 3.2, the national reserve margins (obtained here by dividing total dispatchable net capacities by peak-load, including net exports) might fall below a comfortable threshold of 15 per cent by 2015 in several Member States. However, the reserve margin should remain high in countries such as Greece or Ireland which experienced critical margins during the last decade. Imports should therefore allow robust reserve margins at the European scale.15 Establishing a regional (at least) reserve margin, even taking into account limited interconnection capacities, would therefore be a better approach to generation adequacy than defining reserve margins at the national level. Second, stress events across neighbouring countries are not perfectly correlated, which means that the surplus capacities will generally not be needed at the same time by the different national systems. Of course, stress events are correlated to a certain degree. For instance, weather correlations between neighbouring countries can result in a similar hourly load among different systems, as well as the correlation of weatherdriven output from intermittent RES among these systems. An analysis of historical levels of capacity margins (without flows through interconnectors) from 2005 to 2011 in GB and several other Member States (France, Ireland, Spain, the Netherlands, and Germany) has shown that there was indeed a correlation between low (less than 20 per cent) hourly capacity margins in GB and low hourly capacity margins in Ireland, as well as a correlation between low hourly capacity margins in GB and low hourly capacity margins in France.16 Yet, there was no correlation between capacity margins in GB and any other system at times of very low capacity margins (less than 10 per cent) in GB. While the majority of stress events with low capacity margins occur in winter, they rarely coincide.17 Similarly, ENTSO-E’s regional analysis of generation adequacy for years 2014–203018 shows that the only block of countries requiring simultaneous imports would be Denmark, Germany, Czech Republic, and Switzerland in winter 2025. Yet, there is then a sufficient import capacity available on the external borders of this group to cover its needs. Therefore, resources can be effectively shared at the regional level to cope with non-coincident stress events. Self-sufficient national 15 THEMA consulting group, Capacity mechanisms in individual markets within the IEM, Report for the Directorate-General Energy of the European Commission, June 2013. 16 Pöyry, Analysis of the correlation of stress periods in the electricity markets in GB and its interconnected systems (Report to OFGEM, 2013). 17 Pöyry (n 16). 18 ENTSO-E, Scenario Outlook and Adequacy Forecast 2014–2030, June 2014 (ENTSO-E’s Report).

50%

8%

32%

25%

12%

25%

61% 43%

58%

57%

58%

59%

33% 30%

35% 35%

22% 15% 10%

15% 4%

17% 27% 33%

18%

34%

32%

72%

65% 30%

21% 12%

25% 12% 40%

53%

45% 93%

16%

31% 18% 4%

78%

18%

15%

58% 38%

59% 47%

54% 21%

22%

13%

Figure 3.2 Reserve margin trends in the short-term in Europe Source: THEMA consulting (n 15).

44%

5%

3.3 Costs of self-sufficient capacity mechanisms

47

policies on generation adequacy will require a higher level of investment in generation capacity than a regional approach. Excluding cross-border resources from generation adequacy policies will therefore lead to higher overall costs, as the potential to share capacity resources on a regional scale is lost. The extra cost when nations explicitly provide for their own security of supply at a Member State level was estimated to be €3.0 to 7.5 billion per year from 2015 to 2030, reducing the benefits of an integrated energy market by more than 20 per cent.19

3.3.1.2 Reminder: generation adequacy policies will also impact energy markets Generation adequacy and flexibility are two cornerstones of a reliable power system hosting a high share of intermittent RES. However, experts sometimes perceive flexibility and adequacy as separate issues: certain long-term mechanisms would be implemented to ensure generation adequacy, while other short-term signals from day-ahead to the balancing horizon would ensure an optimal dispatch of the generation and activation of flexible system components (generation, demand, or storage). However, such a distinction between long-term adequacy building and short-term flexibility activation seems misleading, as explained in section 3.2. The ability of system resources to start-up and ramp-up quickly, to cycle frequently, and to operate at low minimum loads inevitably becomes critical in a system hosting a high share of intermittent RES. For a thermal unit with significant variable costs, being available when needed may imply generating at a loss. Generation adequacy incentives inevitably end up impacting prices and flexibility remuneration in the ‘short-term’ markets (energy market, market for ancillary services). If the generation adequacy policy is biased ex ante towards a given set of resources (by technology and location), further distorted conditions are introduced in the short-term markets acting ex post. This pervasive impact of capacity remuneration on other deliverables will be a source of additional costs.

3.3.2 . . . and will not be accepted by the Commission According to the Treaty, Member States have the right to determine their generation mix and define the general structure of their energy supply.20 Also, on a political level, the Commission is reluctant to intervene and define the conditions and operation of security of supply within national boundaries, which has resulted in a variety of country-specific regulatory frameworks with national balancing arrangements and reliability standards. Finally, Article 8 of the 2009 Electricity Directive provides that Member States may organize tendering for new capacities or any procedure equivalent in terms of transparency and non-discrimination, in the interests of security of 19 Booz & Company, David Newbery, Goran Strbac, Danny Pudjianto, Pierre Noël and Leigh Fisher, Benefits of an integrated European energy market (Report for the Commission, Directorate-General Energy, 2013) p 89. 20 Art 194 TFEU.

48

Capacity Mechanisms in the European Market

supply.21 That being said, it appears that the Commission will not allow national schemes to destroy the integration of the internal energy market by the continued vigorous enforcement of state aid rules, competition policy, or the internal market provisions. The Commission’s approach to capacity mechanisms is discussed in other chapters of this book.22 For the purpose of this chapter, we would only like to mention the Commission’s guidance on state intervention, which states that ‘Member States, when intervening to ensure generation adequacy, should choose the intervention which least distorts cross-border trade and the effective functioning of the internal electricity market.’23

3.4 A framework for coordination of national capacity mechanisms at EU level National capacity mechanisms should take into account the contribution of crossborder resources, either to avoid the extra costs of national self-sufficiency or to ensure compatibility with EU law. In this section, we identify a set of tools required to ensure a minimal level of coordination of national capacity mechanisms.24 The first tool needed is a methodology allowing for an assessment of the adequacy needs and a measurement of the contribution of cross-border resources. The second tool required is a regulatory framework for the legitimate remuneration and allocation of responsibilities for delivery of energy by cross-border resources committed in capacity mechanisms. Finally, the last tool needed is a multilateral framework to coordinate the allocation of rights to consume energy at times of general and multilateral scarcity.

3.4.1 A consistent assessment of adequacy needs and measurement of cross-border resources ACER’s Report underlines that ‘the contribution from cross-border capacity to security of supply is often [insufficiently considered] when addressing national or local adequacy concerns.’25 CEER in its report on generation adequacy from 2014 reminds of the remaining difficulties to assess regional security of supply.26 It states that national generation adequacy outlooks are established with no consistent definitions, methods, or scenarios, and in most cases with no identification of the impact of regionally correlated stress events on security of supply. It highlights an urgent need for the harmonization of methodologies within Europe. In that respect, it calls for a 21

Directive 2009/72/EC of the European Parliament and of the Council of 13 July 2009 concerning common rules for the internal market in electricity and repealing Directive 2003/54/EC [2009] OJ L 211/55 (2009 Electricity Directive). 22 See Part III (chapters 9–11) for an in depth analysis of legal issues. 23 Communication from the Commission, Delivering the internal electricity market and making the most of public intervention, C(2013) 7243 final (November 2013 Communication) p 4. 24 We do not evaluate the costs of such a process; that should be compared to the benefits of sharing resources at a regional scale. 25 ACER, Report on capacity remuneration mechanisms and the internal market for electricity, 30 July 2013 (ACER’s Report) p 9. ACER’s Report is discussed in detail in chapter 2. 26 CEER, Assessment of electricity generation adequacy in European countries, March 2014, pp 5–6.

3.4 A framework for coordination at EU level

49

Table 3.1 Harmonization and assessment challenges identified in methodologies for assessment of generation adequacy Type Of Challenge

Parameter

Main Variants

Harmonization

Reliability standard

– No standard – Probabilistic assessment of generation adequacy – Capacity margin – Considering closures due to low profits or not – No reference – N-1 rule – System states (normal/alert/disturbed) – Resource unavailability – Statistical – Model-based – Historical data – Fully unavailable – Percentage of firmness – Detailed modelling – Copperplate approach – Optional transmission constraint analysis – Interconnection modelling (Isolated system or historical data mostly) – No impact of simultaneous severe conditions in different systems – Sometimes weak or no consistency – No strong connection with load forecasts

System stress

Peak-load Assessment

Treatment of variable generation

Network modelling

Correlations Scenario consistency with ENTSO-E scenarios Source: Authors’ own categorization based on CEER (n 26).

more robust and comprehensive methodology to assess security of supply at a regional scale, as the direction and volume of cross-border flows are the result of partially correlated conditions, such as the load and output of intermittent RES in different countries. The main harmonization and assessment challenges that we identified from CEER’s report are presented in Table 3.1. Harmonization challenges are the topics for which satisfactory methodologies exist, but differ widely between Member States. Assessment challenges are the topics for which the methodology could be made more sophisticated in order to properly assess generation adequacy at a regional scale. We identify two ways of including cross-border resources in a (national) capacity mechanism. The first is through explicit cross-border participation, allowing crossborder trade in capacity rights. In other words, foreign generators can participate directly in the capacity mechanism of a neighbouring country on the same (nondiscriminatory) conditions as domestic generators. The second is through implicit cross-border participation, which does not allow for cross-border trade in capacity rights. Instead, it only takes into account the statistical (hence, implicit) contribution of imports in assessing national generation adequacy.27

27

For an in-depth discussion of these two methods see chapter 6.

50

Capacity Mechanisms in the European Market

Explicit cross-border participation is more demanding and has been considered too difficult to implement in France, Italy, or the UK.28 For instance, the UK’s Department of Energy and Climate Change (DECC) clearly states that the government ‘is keen to find a way for interconnected capacity to be able to participate [explicitly] in the capacity market.’ However, for the time being, the UK capacity mechanism will be based on implicit cross-border participation. This means that the expected contribution from interconnection at times of stress will be withdrawn from the amount of capacity auctioned in the capacity mechanism.29 The current evidence points to an average contribution of interconnectors equal to 50 per cent of their nameplate capacity.30 According to DECC, moving to explicit cross-border participation requires finding a practical solution to ensure physical delivery of contracted cross-border capacity.31 Furthermore, the capacity mechanism developed in France only allows for implicit cross-border participation. This is done by employing a multiplying coefficient (‘coefficient de sécurité’) initially equal to 0.93, but revised annually.32 The size of the obligation is equal to the estimated peak demand multiplied by this coefficient. Réseau de Transport d’Électricité (RTE), the French transmission system operator, plans to further improve the participation of cross-border capacity.33 Explicit cross-border participation does exist, for instance, in the Irish Single Electricity Market (SEM). In the SEM, resources receive a capacity payment that is allocated to them on an annual basis according to their availability at times when there is a high loss of load probability. Energy flows across interconnectors also receive half hourly payments: all imports to the SEM receive the capacity payment, while traders exporting energy from the SEM must pay the corresponding capacity payment. However, these resources are remunerated for the production of energy rather than availability.34 On the continent, market coupling makes it difficult to identify the resources that are imported and exported. The mark-ups in SEM, for example, include a component calculated ex-post, which could not be implemented in a market coupling model, as it is in contradiction with the implicit allocation of transmission capacity.35 The implicit cross-border participation, without cross-border trade in capacity, is therefore the only easy option currently available in coupled electricity markets. Yet, implementing implicit cross-border participation is not simple either. Even a basic assessment of the overall contribution of cross-border resources remains challenging. Indeed, despite the high level of interconnectors’ availability (measured by availability factors), the availability of actual energy flows through interconnectors 28

See discussions in chapters 14, 17, and 22. DECC, Electricity market reform: capacity market—detailed design proposals, 2013 (Detailed Design Proposals). 30 DECC, EMR panel of technical experts final report on national grid’s electricity capacity report, 2014, p 21. 31 DECC, Detailed design proposals (n 29) p 23. For a discussion on the UK capacity mechanism see chapter 22. 32 RTE (n 11). 33 RTE (n 11) pp 214–29. For an in-depth discussion see chapter 14. 34 NERA Economic Consulting, The capacity remuneration mechanism in the SEM (study prepared for Viridian, 4 April 2014) pp 4–5. 35 NERA Economic Consulting (n 34) p 57. 29

3.4 A framework for coordination at EU level

51

cannot be easily guaranteed until real time. The potential contribution of interconnectors to the generation adequacy of a given system might actually be highly variable and influenced by concomitant conditions across several European power systems, making it difficult to foresee what can be gained from the interconnectors at times of stress. This is the conclusion of a recent study by Pöyry, which analyses the correlation of stress periods in the electricity markets in GB and interconnected systems.36 The study shows that the historical net interconnector flow to GB has not been driven by system parameters in GB, or in any of the other interconnected systems. For instance, while net flows on interconnectors reduced the number of hours with a very low (less than 10 per cent) capacity margin in 2005, they increased the number of hours with a very low capacity margin in 2007.37 In the absence of a sophisticated probabilistic methodology, Member States might have to exclude the participation of cross-border resources in their capacity mechanisms if these resources already participate (and receive remuneration) in their domestic capacity mechanisms, a policy which is referred to as no double-counting. However, even in neighbouring systems, scarcity events are rarely concomitant. Some resources might actually contribute to generation adequacy in several systems, and at different times. The no double-counting policy therefore overestimates the probability of concomitant stress events in different systems, leading to multilateral overcapacity and extra costs. Mitigating double-counting might come from refining the adequacy products, allowing the time periods of commitments to more accurately match the needs of different systems. The first tool required to ensure a minimal level of coordination between national capacity mechanisms is therefore a methodology sophisticated enough to take into account partially correlated evolutions of load and RES production across different Member States, and a common set of inputs and scenarios shared by the different stakeholders. Such a tool is, in any case, necessary to achieve a regional generation adequacy policy. The emergence of national capacity mechanisms is not a source of new problems, but it does reveal the shortcomings of existing methodologies.

3.4.2 Risk allocation and remuneration of cross-border resources Allocating the risks of non-delivery of committed capacity by cross-border resources is challenging. The actual contribution of committed cross-border resources entails three prerequisites: (a) the resource itself must be available, (b) the interconnector must be physically available, and (c) energy must flow through the interconnector. The risks related to physical availability of the interconnector and resources can be allocated, respectively, to the interconnector operator and the resources operator, who are best able to manage these risks. However, the direction of the flow and the available capacity of the interconnectors are driven by concomitant conditions in different Member States, as illustrated in Pöyry.38 Predicting these market conditions over the long term is problematic, and the associated risk cannot be managed by any entity, in the 36

Pöyry (n 16).

37

Pöyry (n 16) p 23.

38

Pöyry (n 16).

52

Capacity Mechanisms in the European Market

absence of a regional system operator. Allocating the unmanageable risks associated with the direction of flows, to actors who cannot handle them, may have important implications. First, it may result in the reduced participation of cross-border resources. Second, it is likely that national TSOs will tend to be conservative in their assessment of the external contribution to their systems’ generation adequacy. Indeed, this is illustrated by the different levels of reliance on interconnectors expressed in GB by National Grid (0 per cent reliance compared to the nameplate capacity of interconnectors) and the panel of experts consulted by DECC (about 50 per cent reliance compared to the nameplate capacity of interconnectors). Another issue is the remuneration of cross-border resources. The implicit crossborder participation discussed in section 3.4.1 considers the contribution of resources at the regional level, thus avoiding the costs of excess capacity. However, it implies that the contribution of interconnectors and cross-border resources is neither remunerated directly, nor in direct competition with domestic resources. It also suggests that these cross-border resources are not responsible for their contribution to generation adequacy. Remunerating the contribution of domestic resources through a capacity mechanism, while only providing the scarcity rent to cross-border resources, will lead to market distortions. The insufficient remuneration of interconnections could also encourage overinvestment in domestic generation sources pre-empting network infrastructure solutions. Admittedly, the contribution of interconnectors to security of supply is considered in the cost-benefit analysis performed at the time of investment, but it might not be sufficient. As no national entity is capable of managing the risks of non-delivery by the committed cross-border resources, a multilateral regulatory framework aimed at allocating responsibility for the delivery of energy, when needed, will henceforth be a second prerequisite to an efficient contribution of cross-border resources to generation adequacy. Similarly, the network operators and cross-border resources contributing (explicitly or implicitly) to generation adequacy, and exposed to the costs and the risks of delivering energy when needed, should then receive the corresponding remuneration.

3.4.3 Definition of rights over the system resources at times of extreme scarcity The emergence of national capacity mechanisms seeking to insure consumers within a national territory against extreme scarcity can become a source of conflict as Member States do not exist in isolation. The many interdependencies between power systems in Europe make it difficult for a single system (or certain consumers) to ensure their ‘own’ adequacy. This particularly becomes an issue in the case of extreme scarcity, whereby energy prices might not be sufficient to direct energy towards the ‘better’ insured consumers.39 The development of a regional approach to generation adequacy (including solidarity principle) requires the determination of a curtailment level at times of extreme scarcity, ahead of its occurrence. Current rules in the market coupling 39 In this case, price caps that do not reflect the willingness of consumers to pay (insured or not) would be reached in several price zones.

3.4 A framework for coordination at EU level

53

algorithm Euphemia, for instance, impose identical curtailment ratios in all bidding areas, reflecting the principle of solidarity: ‘For those markets that share curtailment, if they are curtailed to a different degree, the markets with the least severe curtailment (by comparison) would help the others reducing their curtailment, so that all the bidding areas in curtailment will end up with identical curtailment ratios in line with all network constraints.’40 It is clear that solidarity at times of scarcity is as important in the electricity sector as it is in the gas sector, and that generation adequacy can be considered a transnational public good.41 Nevertheless, it is also clear that each entity within an interdependent system has to take responsibility to ensure a minimum level of reliability. As some countries (or more precisely, governments) are willing to provide their citizens (voters) with a higher level of insurance against extreme scarcity than their neighbouring countries, it becomes crucial to measure the commitments received by Member States from different resources under heterogeneous generation adequacy policies, in addition to the quantity of energy that consumers in different countries are entitled to, at times of scarcity. Without a common framework to allocate access rights to energy at times of scarcity in an objective and non-discriminatory manner, countries might abandon cross-border socialization (ie free-riding by less insured countries of generation adequacy provided by the ones better insured) and instead, turn towards selfreliance. In simple terms, solidarity principles should not prevent those countries, which paid for a higher level of insurance, from enjoying the higher level of adequacy they contracted for. The commitment of cross-border resources to one country’s capacity mechanism will only be reliable if this country has guaranteed priority access to the contracted resources at times of scarcity. In practice, this means that foreign demand should have priority over domestic demand without a similar contractual commitment. As argued by Pérez-Arriaga (2013), ‘a true security of supply for electricity at EU level will only happen when [physical import and export] contracts have priority over any domestic demand without such contracts.’42 As he further argues, this approach finds its legal basis in Article 4(3) of the 2006 Security of Supply Directive, which states that in emergency situations, Member States should not discriminate between cross-border and national contracts.43 Yet, most national laws contain explicit clauses allowing the interruption of exports in the case of domestic scarcity, which has been applied in practice, despite the Article 4(3) obligation. Member States are also allowed to define cross-border flows after taking into account national flows, while TSOs often have incentives to underestimate cross-border transmission capacity, so as to minimize internal congestion costs. To incentivize the network operator in efficiently allocating

40 Price Coupling of Regions (PCR), EUPHEMIA public description—PCR market coupling algorithm (Version 6.0, 2 October 2013) p 36. 41 For a discussion, see chapter 6. 42 José Ignacio Pérez-Arriaga, ‘Generation capacity adequacy: what economic rationale for support mechanisms?’ (CERRE policy paper, October 2013) p 36. 43 Directive 2005/89/EC of the European Parliament and of the Council of 18 January 2006 concerning measures to safeguard security of electricity supply and infrastructure investment [2006] OJ L 33/22 (Security of Supply Directive) Art 4(3).

54

Capacity Mechanisms in the European Market

the network capacity, the network operator might need to be made liable for security of supply in the neighbouring regions as well. Finon (2013) develops methods to limit capacity leakages, focusing on priority issues between domestic demand with contracts, and cross-border demand without contracts, in an interconnected system. First, he proposes to change the market coupling rules that determine the level of curtailment between interconnected countries in scarcity situations. While current rules tend to equalize the level of curtailment, this level could be adjusted to reflect the efforts made by each country to achieve generation adequacy. A second method, referred to as freezing, is based on the adjusted calculation of available transmission capacity by TSOs when they anticipate scarcity, therefore retaining domestic capacity. Both methods would require an adaptation of the current trading rules on energy exchanges.44 By confronting these arguments, it is clear that the dichotomy between domestic demand and cross-border demand is not sufficient in the context of generation adequacy policies. There is then a clear risk that arrangements allowing country A to rely on imports from cross-border resources in country B (by preventing the withholding of exports from country B at times of joint scarcity) will also, accordingly, prevent country A from retaining the benefits of regional generation adequacy (by preventing the withholding of exports from country A at times of joint scarcity). A regional security of supply will only be possible if the priority of demand with contracted generation adequacy is ensured over demand without contracts, which implies that national generation adequacy policies should be coordinated at a larger regional scale (through bilateral or multilateral agreements between network operators). A first option to allocate rights at times of general scarcity is to provide rules to make sure that the actual level of physical rationing of systems (or their consumers) reflects their efforts made ex-ante to be insured against curtailment. A second option is to introduce an ex-post financial arrangement, whereby systems (or their consumers), which benefit from the activation of the generation adequacy policy without having contracted for it in advance, will compensate systems (or their consumers) which paid for such a policy, but do not benefit from its activation. In any case, it is necessary that there are consistent agreements on both the volume of resources that systems (or consumers) are entitled to, and the value scale of the requisite financial compensation. The issue returns to the value that is placed on the efforts of certain systems (or consumers) to ensure adequacy at times of high scarcity.

3.5 Conclusion The introduction of capacity mechanisms in a number of EU countries has revived the old debate on energy-only markets versus capacity remuneration that has already been at the centre of extensive economic research. However, this chapter focuses on two issues that are particularly important in the European context. First, as the share of 44 Dominique Finon, ‘Can we reconcile different capacity adequacy policies with an integrated electricity market?’ (CEEM Working Paper 2013–2015, September 2013).

3.5 Conclusion

55

intermittent RES has become significant in many European power systems, a satisfactory capacity reserve margin would not be sufficient to ensure generation adequacy. In addition, an adequate level of flexibility is also needed. Second, cross-border power exchanges and allocation of cross-border transmission capacity are efficiently arranged by market coupling algorithms. Generation adequacy policies should not lead to a costly move away from an internal electricity market. This chapter first explains, in section 3.2, that it is illusory to try and separate longterm signals for resource adequacy from short-term signals for flexibility. In a system featuring a large share of intermittent resources, ensuring generation adequacy is not about securing a certain adequate level of installed capacity, it is about providing an adequate, flexible mix of resources. It implies that even the simplest capacity mechanism will not only remunerate capacity, but also lead to a redistribution of the remuneration for flexibility. Therefore, excluding some resources (by technology or location) from the capacity mechanism will also determine the providers of flexibility. This is one of the reasons why ensuring participation of cross-border resources in national capacity mechanisms is crucial. Section 3.3 of this chapter then argues that while it is unlikely that a single harmonized capacity mechanism design could fit the needs of all Member States, national capacity mechanisms should not be implemented in a vacuum, which would lead to overall excess capacity. Hence, the Commission is determined to use its powers under EU law to mitigate the impact of national capacity mechanisms. Finally, section 3.4 identifies three prerequisites to coordinate national capacity mechanisms at the EU level. First, we need a coherent assessment of the EU and each Member State’s actual generation adequacy (taking into account interdependencies and the contribution of cross-border resources). Secondly, we also need to allocate the risks for non-delivery of the committed cross-border resources, which is particularly challenging in the absence of a regional transmission system operator. Physical delivery of cross-border energy is indeed the result of partially correlated and not perfectly-predictable conditions in different Member States. Finally, solidarity and mutual insurance principles should be reconciled with the willingness of some actors to insure against scarcity at a higher level than their neighbours. For a regional approach towards generation adequacy to emerge, it is necessary to recognize the greater efforts made by these actors when allocating the corresponding rights (physical or financial) to consume energy at times of scarcity. It is clear from the diversity of views shared in this publication that the EU debate on capacity mechanisms is not going to end abruptly, or in the foreseeable future.

PART II ECONOMICS

4 Energy Market Design with Capacity Mechanisms Jens Perner and Christoph Riechmann

4.1 Introduction Governments in various EU countries are increasingly concerned about the ability of liberalized electricity markets to provide a stable investment climate that ensures the sustainable evolution of the sector and ultimately security of supply. Most common national designs in the EU presume that power stations are remunerated for the energy they produce while there is no obligation to also explicitly reward capacity. As already explained in the previous chapters,1 this design is often labelled ‘energy-only market’, even though this can at best be a short-hand description of prevailing regimes. For simplicity, we refer to the term ‘energy-only’ market in the following, but mean the comprehensive arrangements in a country which rewards wholesale power mainly through a per MWh-charge, even though there may at the same time be voluntary capacity related contracts such as back-up or option contracts. Even today some plant is remunerated for its capacity, eg plants that provide ancillary services (voltage and frequency control, for example) to the local TSO or plants that act as back-up reserve for generators’ portfolios. Furthermore, even in today’s energy-only market, there is an obligation on the seller of power through the balancing arrangements to serve contracted load at every point in time. This means that the respective energy supply contracts, and therefore also the energy prices stipulated therein contain at least an implicit capacity element. Some countries such as the UK or France have recently initiated policies that are intended to more explicitly provide additional revenue streams to plant operators for capacity they provide to the market.2 Other countries in Europe such as Spain or Italy had previously used less sophisticated means of remunerating capacity. More sophisticated schemes of capacity remuneration exist in several, but not all states in the USA. A number of factors, varying from country to country, motivate these actions: First, a depletion of historical conventional plant reserve margins within individual countries (eg France) as plant dispatch increasingly follows a transnational optimization via short-term markets. This implies that European market interaction eradicates excess capacity by making the least viable plant exit the market first. The resulting geographic distribution of remaining plants will not follow politicians’ desire to maintain a certain national capacity balance. It will—without further policy intervention—likely result in 1 2

See, for instance, section 1.1. See capacity mechanisms currently implemented in France (chapter 14) and the UK (chapter 22).

60

Energy Market Design with Capacity Mechanisms

some countries exhibiting plant capacity in excess of national peak load and other countries that will rely on imports to serve their peak demand. Secondly, uncertainty over the politically controlled expansion path of subsidized renewable energies adds to average plant capacity and depresses electricity prices (eg Germany). This raises concerns about investment incentives for unsubsidized plant. Thirdly, there is uncertainty about the availability of the growing non-dispatchable renewable generation from wind and solar. This raises concerns about market volatility and possibly also the physical stability of the system. In this chapter, we consider the challenges in designing the rules of the electricity market with increasing shares of (subsidized) renewable plant capacity. In doing so, we also consider to what extent the design of renewable support schemes can contribute to sustainable plant investment signals. This chapter is structured as follows. Section 4.2 looks at the mechanisms in the energy-only market that work in favour of security of supply and the challenges— including potential market or policy failures—that may arise without explicit remuneration of plant capacity. This section is an important base for the further discussion: Economists would only endorse policy interventions if the existing system design exhibited systematic market failures. If the claims of market failures were unfounded then so would be any proposals for policy intervention in the market. We consider some of these proposals in section 4.3, where we discuss how to create efficient capacity investment incentives. Section 4.4 focuses on the challenges that arise from increasing penetration of the market with politically privileged capacities such as from renewable energies. Finally, in section 4.5 we look at the EU dimension of security of supply and consider interactions between differing market design choices in different EU Member States.

4.2 Energy-only market—can it be sustainable? Within the current energy-only market design, the services of electricity generators are remunerated through an energy-based payment (€/MWh) in the first instance. The regime has been criticized by academics (initially mainly from the USA), practitioners (especially also in Latin America), and recently also European policy makers.3 In particular, the regime is alleged to insufficiently reward the provision of reserve capacity from flexible and dispatchable plants4 such as gas-fired power stations or electricity storage plants (eg pumped hydro storage plants), preventing them from operating profitably. This would lead to the premature closure of required reserve capacity and lack of investment incentives for new capacities. Such a development would ultimately and fundamentally threaten the security of electricity supply. 3

For examples from Latin America, see chapter 7. In dispatchable plants the operator can control the output profile and increase or decrease output at his discretion. Thermal power stations such as coal and gas plants are typical examples of dispatchable plant. By contrast, in non-dispatchable plant the output profile is subject to external forces, typically climate conditions. Typical examples of non-dispatchable plant are wind or solar power generation technologies. For these technologies the operator has the discretion to reduce the output, but he cannot increase it to nameplate capacity if there is no wind or solar radiation. 4

4.2 Energy-only market—can it be sustainable?

61

In order to judge whether the market design has systematic failures we explore the possible causes of market failures in several steps. First, we explore the functioning of the energy-only market and ask how plant capacity is being rewarded and if this may be sufficient? Next we explore possible market failures and ask what arguments could provide a theoretical justification for political market interventions

4.2.1 Functioning of the energy-only market In the energy-only market, plant operators will offer their production based on the variable (or short term marginal) cost of their plants (at least in the absence of market power), while consumers or their agents (the retail suppliers) will signal their willingness to pay for electricity as buyers. The result is a wholesale market clearing for very short delivery periods (eg each individual hour) which establishes a uniform price for all electricity traded in relation to that delivery period and for a given location. This mechanism allows producers to earn revenues above their own variable cost (and thus achieve a contribution to fixed and capital costs) in two instances, as we can observe in Figure 4.1.5 The left graph of Figure 4.1 shows that, in any hour, generators with power stations with lower variable cost than the last plant which is required to meet demand, ie the price-setting plant, will earn a margin on their own variable cost as the uniform market price exceeds their own variable cost. In technical economic terms this is known as a producer surplus. The right graph of Figure 4.1 illustrates that generators can earn additional rents in tight capacity situations. This is the case where less plant capacity is available than latent demand,6 leading to scarcity prices. These prices will exceed the variable cost of the most expensive power stations. They will rather reflect the electricity price at which certain consumers are willing to reduce their demand or the long-term marginal costs €

€

Nav

Nav

P1 Producer surplus

GK P0

Scarcity premium

P0

Producer surplus MW

Npeak

MW

Figure 4.1 Functioning of an energy-only electricity market (peak-load pricing theory) Source: Frontier Economics.

5 For further explanations of the functioning of competitive power markets, see Steven Stoft, Power System Economics: Designing Markets for Electricity (New York: IEEE Press and Wiley-Interscience, 2002) or Frontier Economics and FORMAET, ‘Decentralised capacity obligations—a promising alternative to centralised capacity mechanisms?’, Study on behalf of the German Federal Ministry of Economics and Technology (German only; English short version available upon request) (May 2013). 6 By latent demand we mean aggregate demand that would prevail at (very) low prices of electricity.

62

Energy Market Design with Capacity Mechanisms

of power generation including capacity costs. Through this, any generators—including the marginal generator and possibly also consumers who can offer demand flexibility (at lower cost than the marginal consumer)7—will earn a margin on their variable cost. This so-called ‘scarcity premium’ can offer a further contribution to amortize fixed and investment cost. This is known as ‘peak-load pricing theory’. In such a world an energy-only market—without regulatory intervention in the form of capacity mechanisms—can in principle provide sufficient investment signals and secure adequate plant capacity.

4.2.2 Current low electricity margins as a justification for capacity mechanisms? A major argument voiced in favour of introduction of capacity mechanisms in today’s policy debate is that current low electricity wholesale prices would not facilitate plant investments and could even trigger the decommissioning of existing plants. This is a concern for policy makers who intend to further increase the role of non-dispatchable renewable generation (wind, solar), in particular. In this setting, it is however important to clearly identify the cause of current low electricity margins since (temporarily) low margins alone do not justify political intervention in the market. In Figure 4.2 we distinguish four potential reasons for the current situation: Evolving market environment Demand reduction

Potential reasons

Unexpected fuel prices changes Market-driven volatility in CO2 market

Unpredicatable deviation from announced subsidized RES deployment Discrete, unexpected changes in the CO2 regime

Power plants currently not profitable

Symptoms/ Implications

Design options

Political intervention

No actions needed

‘Stranded cost’ compensation?

Temporary financial losses consequence of commercial risk

Also: Creating sustainable regulatory environment (eg binding RES maximum targets)

Suboptimal energyonly market design responsible for RES integration Balancing responsibility Balancing energy prices

Fundamental energy-only market failure, eg in form of Security of supply as public good? Prohibitive price risks? Missing money by policy/regulation?

Generation capacity insufficient in the longterm Amendment of the energy-only market design

Introduction of capacity mechanism

eg balancing obligation on RES, asymmetric imbalance prices, ...

And/or starting from reason of problem (eg demand side management Focus promotion)

Figure 4.2 Arguments discussed in favour of the introduction of capacity mechanisms Source: Frontier Economics.

7

See section 1.2.2 for a discussion on the role of DSR.

of chapter

4.2 Energy-only market—can it be sustainable?

63

4.2.2.1 Evolving market environment First, the evolving market environment has led to a period of under recovery of cost. It should be noted that current low electricity wholesale prices are themselves the result of existing overcapacity in the market and can be readily explained by the workings of market mechanisms. Low energy and electricity demand in Europe are a result of the economic crisis and structural changes that it induced. Some overcapacity in conventional plant was inherited from the pre-liberalized (and in some cases also postliberalized) era. Furthermore, the economic crisis after 2008 has led to low prices for CO2-certificates worsening the commercial viability especially of gas-fired power plants. Finally, the development of fuel prices such as gas prices may deviate from expectations of market participants some years ago. Therefore, investments in gas-fired power plant capacity may be lower than expected. These developments reflect market risks and should not be seen as an economic justification for the introduction of a capacity mechanism. Certain risks should be accepted by market operators.

4.2.2.2 Political intervention Secondly, market investors may have been ‘tricked’ by certain changes in the policy environment. For example, excess capacity was created through the expansion of subsidized renewable capacity (not only in non-dispatchable technologies but also in biofuels etc) and subsidized cogeneration capacities. Furthermore, an investment boom was triggered in conventional plant in connection with the introduction of the European Emission Trading Scheme (EU ETS) that initially saw generous rules for the free allocation of emission permits in the early phase in some Member States. As could be expected, these overcapacities result in low electricity wholesale prices. This alone is not an indication of any market failure. Power generators claiming for a compensation are really asking for ‘stranded cost’ payments to compensate for a specific and historic policy decision. Such circumstances do not justify the introduction of capacity mechanisms, though.

4.2.2.3 Design flaws Thirdly, deficiencies within the specific design of energy-only markets in individual countries (rather than deficiencies in the functioning of energy-only markets per se) may imply that the market does not meet its role of ensuring security of supply through sufficient capacity—at least not in the most efficient way. A number of deficiencies have been discussed in practice. For example, some countries like, eg Germany have initially used very weak price signals in energy balancing regimes. For example imbalance prices were calculated on the basis of average bids rather than marginal bids. This has implied that market players who did not keep a balance between their injection into and their withdrawal from the grid faced no penalty and might have even benefitted from being out of balance. Such a setting clearly does not provide clear incentives for market players to back their contracted sale with capacity as it could be commercially advantageous to simply top up the imbalance from power on the grid. If

64

Energy Market Design with Capacity Mechanisms

all players behaved in this way and the reserve margin became slimmer, the likelihood of outages would rise over time. Further challenges arise in the context of increasing penetration with electricity from renewable energy sources. We discuss these in a separate dedicated section later in the chapter. Market failures that result from an inappropriate design of the energy-only market are best addressed——where possible—by refining the rules of that very energyonly market design, eg by sending clear price signals through the balancing mechanism. This might be achieved, eg through prices based on marginal bids and asymmetric prices for shortfalls and excess supplies. Where it is feasible to resolve the failures within the current regime, the introduction of more far-reaching capacity mechanisms is not required. We return to the options to enhance the current energy-only market design in the third section of this chapter.

4.2.2.4 Market failures Fourthly, there could be genuine and structural failures of how energy-only markets work. If such failures exist, and these failures cannot be healed by less intrusive means, then the introduction of certain capacity mechanisms becomes a viable option. More fundamental changes such as the introduction of capacity mechanisms would then require a different justification: By more fundamental failings even in a welldesigned energy-only market. We discuss these potential fundamental failures in the following section.

4.2.3 Possible fundamental market failures The academic literature8 and the practical debate have identified several possible fundamental causes of market failures, broadly in the three categories of externalities/public goods nature of security of supply given inelastic demand, volatility which may create prohibitive prices risk, and the threat of or the actual political intervention in price formation (missing money) which are presented in Figure 4.3.9 In essence, all these problems may cause plant capacity to be insufficiently remunerated. This in turn would lead to underinvestment and an inefficiently low level of security of supply. Some authors also raise concern about the use or abuse of market power in capacity constrained markets. The issue here, though, is not that of an insufficient remuneration of plant, but of excessive remuneration. However, abuse of market power may lead to 8 See for instance Peter Cramton and Steven Stoft, ‘The convergence of market designs for adequate generating capacity—with special attention to the CAISO’s resource adequacy problem’, A White Paper for California’s Electricity Oversight Board (April 2006); Dominique Finon and Virginie Pignon, ‘Electricity and long-term capacity adequacy: the quest for regulatory mechanism compatible with electricity market’ (2008) Utilities Policy 16 (3), 143–58; William W. Hogan, ‘On an “energy only” electricity market design for resource adequacy’ (Working paper, Center for Business and Government, Harvard University, 2005). Pablo Rodilla and Carlos Batlle, ‘Security of electricity supply at the generation level: problem analysis’ (2012) Energy Policy 40 (C), 177–85. 9 See also Frontier Economics, Formaet Services, ‘Electricity market in Germany—does the current market design provide security of supply?’ (Study for the BMWi, September 2014).

4.2 Energy-only market—can it be sustainable?

€

Inelastic demand

Uncertainty and volatility

No market clearing

€

Regulation €

Price forecast difficult

 Blackouts/ rationing!

65

missing money

Npeak

Npeak

Npeak

MW

MW

MW

Figure 4.3 Potential fundamental market failures in an energy-only market Source: Frontier Economics.

political intervention in price formation and therefore indirectly to the ‘missing money’ problem stated earlier. In relatively small and insular markets (eg Ireland), there may be a further concern about the effects of lumpy investment. While market prices may be sufficiently high before a new investment, the efficient scale of a new investment may be so large that for some time it depresses the market price. The effect may be that no single player is ever prepared to invest.

4.2.3.1 Security of supply as ‘public good’ The academic literature takes particular concern with the combination of two features: First, consumers cannot or do not express their willingness to pay through the market, respond insufficiently to price signals (consumers are price-inelastic). Secondly, in situations with actual generation shortage and partial power disruption in defined parts of the grid areas (brownouts), producers located in the affected area cannot earn any revenues or margins. Furthermore, in a situation with partial disruption (brownout) the prevailing energy price may not reflect the social value of electricity to consumers as some demand may be curtailed and the market price has to be defined administratively. This in turn has two consequences: First, power generators, which are available to produce in a brownout situation may only benefit from electricity prices administratively defined for the hours in which demand is technically curtailed. If these prices turn out not to reflect scarcity prices, producers will reflect this lower valuation in their investment (and disinvestment) decisions. As a result of this externality, less generation capacity may be held on the system than is economically and socially desirable. The provision of (reserve) capacity has characteristics of a public good. Secondly, vis-à-vis consumers there may be inefficient rationing of demand in case of power supply failures. When demand responds insufficiently to high electricity prices in periods of a capacity shortage, random rationing of demand can occur (blackouts or brownouts). Some consumers, who value reliable power supply more than other

66

Energy Market Design with Capacity Mechanisms

customers, will not have the opportunity to effectively signal their higher valuation, and they may be interrupted along with other consumers in their network area. This insufficient demand response may be caused by a number of factors. Institutional barriers may limit the implementation of time-of-use or real-time electricity tariffs. For example, in some countries, while smart metering technologies exist that allow for real-time pricing, respective tariffs may not have been approved by authorities and can therefore not be applied. Furthermore, technical and commercial conditions (such as high cost of the required consumption metering technology or inadequate consumer interest in required tariffs and metering technologies) may limit the demand response by consumers. These circumstances could potentially justify a policy intervention in the market design to achieve the socially desirable level of capacity on the system. However, the measures should be selected very carefully. These could range from weak interventions such as defining rules for price determination and settlement for hours in which demand is technically partly curtailed to the implementation of comprehensive and complex capacity mechanisms.

4.2.3.2 Underinvestment due to prohibitive price risk (uncertainty for investors) If plant investors depend on scarcity rents to amortize their investment cost, they will consider investments as particularly risky. If, on the other hand, there is uncertainty about the frequency and level of price spikes, risk averse investors will become more cautious and will require higher risk premia, either to invest or to continue operation. In both cases, this will eventually lead to higher prices for consumers. This situation can become particularly relevant as the volatility of electricity wholesale prices rises with the increasing penetration of wind and solar generation. The basic argument here is one of a market failure in financial markets (to fund the continued operation of power stations) and not one of a failure in the electricity market itself. The issue may be exacerbated if uncertainty over the exact degree of political promotion of new generation technologies such as renewables prevails. We address this issue in a separate section later, but focus just on the issue of market driven volatility for now. Under certain circumstances this may justify policy intervention in the market by mandating capacity mechanisms. This may be the case if the social benefit of risk reallocation (where part of the risk is shifted from plant investors to the collective of consumers) outweighed the detrimental effects of the intervention, eg less reliance on dynamic and market-based solutions, increased regulatory and policy risk. However, it should be noted that capacity mechanisms are not the only way to handle some of the aforementioned risks. The electricity market offers opportunities to limit or reallocate risks via forward and option contracts, some of them already prevailing today. Moreover, the signal of increased market risk may be an important one. As a result of increasing renewable generation the fundamental risks10 in the electricity market are 10 The term fundamental risk refers to risk which is related to certain underlying and often physical factors in the market, such as the level of demand, the availability of wind and solar radiation, or fuel prices. Wind and solar availabilities are fundamental factors that have in many countries only achieved significance over the last decade.

4.3 What are the policy options—and how would they perform?

67

increasing and therefore it may be economically efficient for investors to consider them in their investment decisions by increasing risk premia and return requirements. In this sense it is questionable whether the increase in market risk alone justifies policy intervention in the market design by setting up capacity mechanisms. We address policy risk related to the extent of renewable support in section 4.4 below.

4.2.3.3 The threat of exercise of market power in peaking periods The next market failure consists not in energy prices being too low to reward risky plant investment, but rather in prices being too high. The more scarce plant capacity is, the easier it can be for dominant firms to exploit their market power by raising prices in scarcity situations to the detriment of consumers and overall welfare. An effective capacity mechanism should lead to additional plant capacity and fewer scarcity situations and as such, mitigate the risk of market power. Moreover, capacity mechanisms can be designed to explicitly limit the price at which generators with a capacity contract can offer their output in the market. However, we would like to note again that the introduction of a capacity mechanism is not the only remedy that may be considered here. Effective antitrust enforcement in case of market abuses may yield equal (or higher) benefits. Furthermore, market power can also exist in capacity mechanisms. Antitrust enforcement in case of market abuses may be even easier in an energy market than in a capacity mechanism since the economic rational of investment and disinvestment decisions may be more difficult to assess by cartel authorities in a capacity mechanism. Relatively high prices may also be desirable to induce new investment and attract new entrants, although there may be a temporary risk of high producers’ rents and higher than optimal prices for consumers as a side effect.

4.2.3.4 Missing money problem due to threat of regulatory intervention In regulatory practice, it is difficult to distinguish between an appropriate price spike necessary to signal scarcity and a high price as a result of abuse of market power. For this reason, as explained in section 1.1, regulators often introduce price caps, which lead to lower revenues for generators to cover their fixed and investment costs. This is known as the missing money problem, which might justify the introduction of a capacity mechanism in order to compensate for such lost revenues. Even the sole threat of introducing price caps may lower investors’ return expectations and even this might need to be addressed, if only by policy makers reassuring the market that prices would not be capped.

4.3 What are the policy options—and how would they perform? 4.3.1 Reform without capacity mechanism As described earlier, an analysis of the energy-only market suggests that some deficiencies may prevail in the current market designs in some countries. Before embarking on the design of capacity mechanisms it is therefore reasonable to consider the root

68

Energy Market Design with Capacity Mechanisms

cause of these flaws in the design. We explore three potential causes and corresponding remedies. First, enhancing the flexibility of demand and integrating demand response in the electricity market would help remedy an important shortcoming in the current regime. Sophisticated metering and billing systems are available and would be required to mobilize demand response (and they have already been installed in some Member States). For this to be effective it would not even be required to cover the entire market demand. Initially, it would be sufficient to focus on large consumers such as industry and commerce. However, enhancing the flexibility of demand and integrating demand response into the electricity market may also imply additional costs which have to be taken into account when assessing such options. Secondly, stability of the policy framework is another ingredient to a wellfunctioning market. This requires a clear policy commitment of how policy makers and authorities will respond to high electricity prices in scarcity situations, as well as clarity over whether capacity mechanisms will be implemented. Uncertainty over the introduction of capacity mechanisms in the future may reduce investment incentives today. Furthermore, keeping renewable support to announced renewable expansion targets falls into this category. Thirdly, ensuring market liquidity is important. In markets where liquidity has not matured and therefore where a market lacks reliable medium-term price signals and market responses to extreme events, measures may be sought to enhance market liquidity. This could involve structural changes such as increasing cross-border capacities or softer behavioural remedies such as obliging key players to act as market makers.11 Further options to enhance the current market design could entail a reform of the balancing energy regime. As far as it does not exist already, introducing marginal pricing with high prices in case of demand exceeding the schedule and low prices when demand is below schedule can create commercial incentives for market players to hold required reserve capacities. It will also be useful to introduce rules for settlement prices to be applied in situations on scarcity and brownouts in the extreme. In such circumstances settlement prices should ideally reflect the very high valuation of power by consumers that are actually or potentially interrupted.

4.3.2 Which capacity mechanisms are debated? The next level of escalating policy intervention in the market would be to consider explicit capacity support mechanisms. Several distinct options have emerged in the international debate, including strategic reserve, capacity auctions, capacity obligations, reliability options, and capacity payments, and we refer the reader to section 1.2.3 11 A market maker is a market participant who will continuously and simultaneously offer to sell and to buy a certain volume of power. By offering to buy and sell at the same time, market participants will always be able to find a counter-party in the market. The market maker will be selling the power at a slightly higher price than he offers to buy. The difference between these prices is known as the (bid/ask) spread. A low spread suggests a very liquid market.

4.3 What are the policy options—and how would they perform?

69

for an overview of these types.12 In this chapter, we focus on capacity mechanisms where capacity price is determined through some sort of competitive mechanism13 and we classify them in several dimensions. In terms of centralization we can distinguish whether capacity is procured by central agents (eg the TSO or a state authority in the case of reliability options or the strategic reserve) or by decentralized agents (eg in the case of capacity obligations on retail suppliers). In terms of reach and scope of the mechanism, we can distinguish whether all capacity on the system benefits from payments from a capacity mechanism (eg in the case of reliability options or capacity obligations) or only certain qualifying plant. In terms of continued participation of plant in the energy market, we can distinguish whether plants (benefitting from capacity mechanisms) can freely participate in the energy market or whether they are excluded from it. In the case of a strategic reserve, contracted plants are purposely excluded from the energy market (partly to ensure that the introduction of capacity contracts does not significantly distort price formation in the energy market). Under reliability options plants can freely participate in the electricity market. However, in case the electricity spot price exceeds the strike price in the option contract, electricity committed under this contract is offered at the strike price. If all or most capacities are covered by such option contracts, the strike prices in these contracts may effectively cap the electricity spot price in the energy market in many periods. The implicit price cap would be exceeded only in periods of very high demand, for instance when expensive demand side options are called at very high prices. Figure 4.4 maps the design options discussed in section 1.2.3 in the dimension of centralization and scope/reach.

4.3.3 Which design options address which market failures? When deciding on an appropriate capacity mechanism for any country, those models that allow achieving the required level of security of supply, and at the same time limit the degree of policy intervention into the market would be preferred. Depending on the design choice, the introduction of a capacity mechanism can have significant repercussions on the energy market and revenue prospects of power plants from this market. For example, the impact of a well-designed strategic reserve on the energy market will be limited (if any impact exists) to the few hours per year when scarcity prices trigger See Frontier Economics, Consentec, ‘Impact assessment of capacity mechanisms’, Study on behalf of the German Federal Ministry of Economics and Technology (English short version) (May 2013). The list of possible design options is not exhaustive, and each option can have further variants. For instance, ÖkoInstitut/LBD/Raue have developed a variant for Germany where only part of the capacity requirement is procured by a central agency (eg a TSO). This ‘partial’ capacity mechanism is guided by the distributional consideration of limiting generators’ profits. In this model, capacity payments are paid out only to those plants that would not operate without such payments. All plants continue to participate in the energy market. Please note that, in contrast to other types of capacity mechanisms discussed in this chapter, this particular model has never been implemented in practice. See Öko-Institut, LBD and Raue, ‘Fokussierte Kapazitätsmärkte. Ein neues Marktdesign für den Übergang zu einem neuen Energiesystem’ (in German only), Study on behalf of WWF Deutschland (October 2012). 13 Capacity payments have the disadvantage that they do not directly allow control of how much plant is added to the system. There is a risk of uncontrolled overcapacity and high resulting cost. We therefore do not consider this approach further. 12

70

Energy Market Design with Capacity Mechanisms Level of centralization of the capacity mechanism 1

Scope of the included capacity

Strategic reserve (eg Sweden, Finland)

Capacity not (directly) offered in energy market

2 Partial capacity market (eg LBD / Öko-Institut) 5

4

Decentralized performance obligation for • suppliers (eg FR) • generators

Decentralized obligation + auction (eg PJM)

Funding of capacity via competitive market

3 System-wide centralized reliability option (eg Columbia)

Capacity still offered in energy market

Funding of capacity via administrative charge

Figure 4.4 Classification of capacity mechanisms Source: Frontier Economics. Note that capacity mechanism no 2 is theoretical and has not been implemented in practice (n 12).

its activation (see no 1 in Figure 4.5), depending on the rule by which electricity wholesale prices are set in these specific hours.14 Thus, in certain cases, a strategic reserve might have an equivalent effect to a price cap, which could ultimately lead to distortions in the energy market, however, distortions would be limited to a few hours per year. A reserve mechanism can also be designed such that the price in hours in which the reserve is called is set above the marginal cost of the highest bids on the system (this variant is not drawn in Figure 4.5). By contrast, capacity mechanisms with a broader scope/reach (nos 2, 3, 4, and 5 in Figure 4.4 and Figure 4.5) will also impact on energy prices in mid-merit and possibly even base-load periods. This is because all capacity benefitting from capacity payments is also allowed to participate in the energy market. If the capacity mechanism is effective in securing additional plant capacity, then more capacity will be available and some of this may be new and so more efficient than older plant, thereby exerting a downward pressure on the energy price. The interdependency between the capacity mechanism and the energy market may also result in a ‘waterbed effect’. This can arise if power plants generate extra revenue through capacity mechanisms, but lose revenue in the energy market (see the shaded area in Figure 4.5) as a consequence. While in the aggregate plants may tend to have higher overall revenues, some plants may end up with lower revenues. 14 Apart from that, the strategic reserve can deprive the energy market of certain capacity if it includes plants which otherwise would not have been decommissioned but would be dispatched in the energy market at least for some hours. Excluding this capacity could lead to higher prices in the energy market, however, we expect that this effect is negligible as long as the strategic reserve corresponds to a small percentage of peak demand.

4.3 What are the policy options—and how would they perform?

71

€/MWh 1 Stragegic reserve leads to capping of extreme price peaks through fixed strike price (eg €3,000/MWh)

2 3 4 5

Capacity mechanisms reduce prices in all hours in the direction of variable generation cost of efficient plant (no peak prices, reflective of cost of load reduction)

Shaded area = reduced energy revenue for plants

Peak-load

Mid-load

Price duration curve in energy-only market

Base-load

h

Figure 4.5 Impact of capacity mechanisms on the energy market (schematic illustration) Source: Frontier Economics. Note that prices are shown as a price duration curve, ie hourly prices are ordered by their level starting from high to low prices.

Furthermore, all capacity mechanisms imply additional costs to the system. Beside transaction cost, which is associated with any change in design, there might be additional costs resulting from (a) design flaws and (b) policy failures.15 With respect to design flaws (a), even a well-intended design is likely to exhibit deficiencies. These may ultimately be evened out over time as experience with the performance of a particular design grows. The practical experience in implementing capacity mechanisms shows that these models take years if not decades to converge on a stable design. For example, some designs have experienced volatile capacity prices and have been ineffective at triggering the required capacities.16 Regarding policy failures (b), policy makers may want to use (or misuse) capacity mechanisms for objectives other than those originally intended. For example, for political reasons they may favour certain technologies or types of players over others. Any such policy failures will raise the cost of the design. Therefore, it is important that any decision about the introduction and choice of capacity mechanism is accompanied by an impact assessment: If a market failure (in relation to security of supply) is diagnosed with certainty and it is persistent in nature, the introduction of a capacity mechanism broad in scope, like reliability options or 15 For examples see Fernando Barrera, Matthias Janssen, and Christoph Riechmann, ‘Kapazitätsmärkte: aus der internationalen Praxis lernen?’, Energiewirtschaftliche Tagesfragen no 9/2011 vol 61 (September 2009) pp 8–12. 16 For instance, the Capacity Credit Market as operated in the PJM region of the US between 1999 and 2007 led to highly volatile capacity prices, especially due to an inelastic demand for capacity credits. This design was replaced from 2007 by the reliability pricing model which incorporated an artificial demand curve (known as variable resource requirement) that mimicked some demand flexibility. Early capacity market designs in PJM also implied that capacity was not always available when needed.

72

Energy Market Design with Capacity Mechanisms

capacity obligations, may be justified. In case of a fundamental market failure a partial capacity mechanism (no 2 in Figure 4.4)17 will not provide a sustainable solution. Rather, it would need to be increasingly broadened in scope so that it ultimately covers the entire market. To the contrary, if the market failure is uncertain, temporary or limited in scope, a bridging measure in the form of a partial capacity mechanism might be sufficient. In case of a more severe but still transitory failure, a more comprehensive mechanism may be required. Nevertheless, in such case, the mechanism should also be transitory (sunset clause) or reversible, to keep intervention at minimum.

4.4 How do renewable energies fit into the market design? Energy policy in many EU countries has recently focused on the expansion of RES. Non-dispatchable technologies18 of wind and solar play a particular role in this. Therefore, RES expansion can raise particular issues in relation to future security of supply especially in relation to (a) volatility of generation and (b) inaccuracies of production forecast. In relation to (a), volatile and fluctuating generation from wind and solar requires the availability of secured back-up capacities through storage, conventional generation, or demand flexibility that can step in for these renewables. The question is whether market mechanisms alone can deliver these capacities. In relation to (b) inaccuracies in production forecast, the availability of wind and solar radiation can only be predicted with limited accuracy. The availability of respective plants is only known with high certainty shortly (a few hours or less) before real-time dispatch. This holds even though there have been significant improvements in forecasting accuracy in recent years. This uncertainty about actual production requires back-up capacity that needs to be dispatchable at short notice. Therefore, even absent the potential wider market failures discussed in section 4.2.3, the question arises whether the development of RES raises further issues for security of supply on the system. In the following, we focus on two specific concerns. First, exempting RES from balancing obligations may raise specific requirements for capacity reserves and may lead to inefficient solutions. Secondly, unpredictable patterns of support granted to RES may lower planning certainty for non-subsidized plants and lead to underinvestment in reserves.

4.4.1 Balancing obligation for RES In principle, back-up reserve for RES could be procured centrally or de-centrally. In the case of central procurement, RES producers—unlike conventional producers—would not be under an obligation to forecast their own generation and to match their actual production to their forecast. In order to balance the system, a regulated entity (eg a TSO) would then need to procure and dispatch a back-up reserve for the renewables 17

See n 12.

18

Generation from respective technologies can be regulated down, but not up.

4.4 How do renewable energies fit into the market design?

73

centrally. In case of de-central procurement renewable generators would carry the same responsibilities to provide generation forecasts and comply with them just as conventional generators. This would provide them with the same commercial incentives to procure back-up reserves as conventional generators have at least when selling electricity on the forward market (instead of the day-ahead or intraday market). The model based on central procurement of the reserve has been adopted in most Member States. A number of justifications were provided historically: Portfolio effects imply that it makes sense to secure the back-up reserves for a larger portfolio (eg all renewable capacity connected to the grid of a specific TSO). Moreover, the degree of RES penetration was initially low and so the reserve requirement did not pose a major burden on the TSO. Recent discussion reveals that these arguments no longer apply in the same way. RES today has become a major share of power generation in most EU countries and the production swings and the reserve requirements they impose are significant. Beyond that, given the high degree of RES penetration, the portfolio effect no longer requires all back-ups for renewables to be procured by one agent. Instead, the portfolio benefits can also be achieved by several, potentially competing agents. The historic central procurement model is also exhibiting some weaknesses and inefficiencies: In the first instance, the procurement by the TSO misses some efficiency potentials. Namely, the TSO takes over responsibility for RES and procures it according to legally sanctioned rules and processes. The TSO develops a renewable generation forecast, and then makes up any difference between forecast and outturn renewable production by procuring and dispatching balancing energy according to rules set by the regulator. The legal framework often prescribes that this balancing capacity can only be contracted at short notice (eg on a weekly or monthly basis) so that the incentives and also the remuneration for such back-up capacities differs from that which evolves with respect to conventional generation. This stifles innovation in terms of reserve products, their term structure etc. For example, if longerterm contracts to procure balancing capacity were allowed a different mix of back-up plant may evolve. In the second instance, exempting RES from balancing responsibility leads to missed opportunities. Smart solutions to balance volatile output from renewables may be overlooked. RES generators have no commercial incentives to explore alternative ways to improve the steadiness and predictability of their output. This includes, for instance, diversifying their portfolio by technology or location, using local or central storage options or developing technologies to enhance the ability to control the output. These deficiencies could be addressed by assigning a marketing responsibility to renewable generators that would implicitly also convey a balancing responsibility. For instance, placing RES generators under a direct marketing obligation and promoting them through an additional premium, fixed per kWh, would expose them to market (price) risk. This in turn would create incentives for them to sell their output in forward markets and to search for the most efficient ways to procure a back-up reserve to match their reduced reliability. They could also contract reserve capacity for periods of their unavailability, as a result raising market demand for capacity contracts even without any other policy intervention in favour of capacity mechanisms.

74

Energy Market Design with Capacity Mechanisms

4.4.2 What is the role of the extent of renewable support? A further challenge is that renewable support schemes have led to a rapid expansion of RES capacity at a time when excess capacity already prevailed on the system. As long as the expansion of subsidized RES production is faster than the market requires, electricity wholesale prices will be (further) depressed to the extent that the full cost of existing or potential new non-privileged generation capacity will not be covered without further market interventions including new capacity to ensure security of supply. This creates (a) efficiency issues and (b) distributional and stranded cost issues. With respect to (a) the efficiency issue, falling energy prices induced by rapid renewable penetration will lead to the early exit of the least profitable conventional plants. Some of these exiting plants would have been profitable, if the particularly rapid expansion of subsidized renewables had not taken place. This leads to the destruction of otherwise productive capital (subsidized renewables crowd out otherwise profitable conventional plants). With respect to (b) the stranded cost issue, distributional concerns may arise as conventional plant investments may have been historically undertaken in anticipation of a particular policy environment. Practice shows that—eg for Germany— policies have for some time supported more renewables than had been set out initially as the formulated political ambition. This additional RES capacity (above and beyond earlier announcements) contributes to further overcapacity and creates further downward pressure on energy prices and revenues of conventional plant investors. As a result of this deviation from previously announced policy, investors are unexpectedly unable to cover the full cost of their investment and may need to forego part of it. In order to secure adequate capacity reserves, policy makers now consider additional investment incentives through the capacity mechanisms discussed in this chapter. Even if such capacity mechanisms were implemented, existing investors should be alerted: Such a mechanism does not guarantee the viability of all existing plants, but—if designed efficiently—only of those plants which are really required from a security of supply perspective. Moreover, policy makers should make sure that the expansion of RES production is only subsidized within a clearly committed corridor. At the same time, the expansion of renewable capacity that does not enjoy subsidies should not be constrained. Otherwise, investors who do not benefit from renewable subsidies (conventional plants, but also renewable investors that do not receive subsidies) will continue to be exposed to policy risks that translate into commercial risks. This might create a political need for high capacity payments to compensate for such risks, and can also lead to inadequate capacity reserves.

4.4.3 What is the role of volatile, non-dispatchable renewable generation? A further challenge may arise out of limited ability to forecast output from volatile wind and solar production, imposing further requirements for the flexibility of the generation system. To the extent that this flexibility cannot be procured through existing products in the traditional energy markets (eg day-ahead and intraday), due to, for instance, specific flexibility requirements below the usual fifteen minute

4.5 What are the challenges in the EU context?

75

settlement intervals or particularly steep gradient requirements of generation such products must be procured by the TSO in a separate, balancing market. As a result, the current balancing energy products may need to be further developed and refined. Such refinements could relate to the lead times of contracting, contract duration, lead times for dispatch, and duration of dispatch. To the extent that current institutional arrangements prohibit the TSOs from developing such refinements, regulations may need to be reformed.

4.5 What are the challenges in the EU context? In this chapter we have so far presumed a geographically integrated market with one set of rules. In reality, Europe’s energy markets are sometimes national and sometimes integrated across borders to the extent that available cross-border capacity allows this. Nonetheless, the Commission has invested great efforts to ensure that market rules between different countries are aligned to facilitate cross-border trade. Capacity mechanisms introduce a further complexity as different countries could then opt for different market designs, potentially undermining EU market integration. A respective discussion is only emerging and therefore—at this stage—we can only point out some of the potential frictions in the debate.

4.5.1 Issues in the EU dimension Some of the issues include (a) objectives and targets of the regime, (b) support across borders, and (c) the interaction of capacity mechanisms in different countries. As for (a) objectives and targets of the regime, it is inconceivable that neighbouring countries pursue very different levels of security of supply, given that electricity markets are interconnected and mutually dependent. Therefore some cross-border coordination is required. In relation to the policy objectives, questions include how security of supply should be defined. For example, should its definition be based on output, for instance, as an outage probability, or input, eg in terms of a certain margin of capacity over demand? Once the objective is clear, the question arises which target level should be achieved (high/medium/low), and whether there should be a national, regional, or EU-wide target. Where the target is not EU-wide, should it (and could it) be identical or differentiated per region or country? As for (b), the consideration of cross-border support, the question arises to what extent one country is allowed to rely on reserves of another country and to what extent interconnectors should be considered as a constraint in this. To what extent is one country obliged to help another if an emergency situation arises? Would that imply that some reserve held in country A cannot actually count as reserve for that country? As for (c), the interaction between capacity mechanisms in different countries— different capacity mechanisms could interact in complex ways. Network externalities may lead to market distortions. For example, if country A introduces a mechanism and country B does not, then ceteris paribus new plant investment would now be more attractive in country A. As long as sufficient cross-border capacity is available, consumers in country A can fund a reserve that may also benefit country B. However, if

76

Energy Market Design with Capacity Mechanisms

cross-border congestion occurs or country A tries to reserve local capacity for local consumption, then country B may be left with a capacity gap in the long run especially if the rules for using generation capacity cross-border in scarcity periods are not well defined. As is clear from this simple example, the interaction of national energy regimes can have complex allocative and distributional implications. The need for cross-border participation raises complexity. In all of this, at minimum it should be ensured that plant capacity of country A can contribute to and participate in the capacity mechanism in country B. It must also be ensured that interconnection capacity between A and B is available and that the contribution from abroad really takes place in periods with scarcity. Ensuring non-discrimination between domestic and cross-border capacity may raise the complexity of each capacity mechanism design.

4.5.2 Guidance by the Commission The Commission is providing guidance on which capacity mechanism it would consider admissible. However, it is largely bound by EU law and state aid rules in particular. These however effectively only allow the development of a ‘negative list’ of measures which would be seen in contradiction to the rules of the internal market.19 Figure 4.6 illustrates that any mechanism must follow a well-defined objective of common interest (Article 107 TFEU), must remedy a specific market failure, must be appropriate for this purpose, change behaviour accordingly, must be proportional to the issue, and must avoid unnecessary competition effects.20 This would still leave many options open to the Member States considering setting up capacity mechanisms. The risk is that individual Member States adopt capacity mechanisms which are individually admissible, but which between them are incoherent and inconsistent. In this case market distortions could arise, even if each individual market design as such would be acceptable. This means that a European or at least regional concept of capacity mechanism design may be preferable over the co-existence of several incoherent regimes. Achieving this may require further EU legislation.

4.6 Summary—when to use which capacity mechanisms? Our discussion shows that security of supply may be addressed through various reform measures which may lie within the current energy-only market design or which may expand the design by capacity mechanisms. The remedy—and with it the intensity of market intervention—should be adapted to the cause diagnosed through logical analysis. From an economic perspective, less interventionist measures are preferred provided they suffice to cure the issue. Therefore, reforms of the existing market design 19

For an assessment of capacity mechanisms under the state aid rules see chapter 9. See Commission staff working document, Generation Adequacy in the internal electricity market— guidance on public interventions, 5 November 2013 (Generation Adequacy SWD) and Communication from the Commission, Guidelines on State aid for environmental protection and energy 2014–2020 [2014] OJ C 200/1–55 (EEAG 2014–2020). For a discussion on the Generation Adequacy SWD, see section 1.5. For a discussion on the EEAG 2014–2020, see section 9.4. 20

4.6 Summary—when to use which capacity mechanisms? a

Well-defined objective of common interest (Art 107 (3) TFEU)

b

Need to intervene: Remedying a market failure

c

Appropriateness of measure

d

Incentive effect: Measure should change behaviour of undertakings

e

Proportionality: Minimum amount needed to incentivize behaviour

f

Avoid undue negative effects on competition and trade

The Commission’s presumption

77

Certain categories may be subject to ex-post evaluation. The Commission may limit duration of schemes (indication: four years) and decide on prolongation.

(155) (...) mechanisms that ensure that certain generation adequacy levels are met at the national level may constitute State aid in the meaning of Art. 107 (1) TFEU. (...)

Figure 4.6 The Commission’s guidance on capacity mechanisms Source: Frontier Economics based on the Commission’s EEAG 2014–2020 (n 20).

should be exploited first before embarking on the introduction of more extensive mechanisms in the form of capacity mechanisms.

4.6.1 Reforms of the existing market design The current market design already offers some option for reform. For example, demand flexibility could be strengthened so that in a situation of scarce generation capacity consumers can adjust their demand depending on their willingness to pay for electricity. This helps to overcome some of the potential market failures identified in the literature and it thereby reduces the risk of unplanned and partial or complete power failures. Our discussion on the (inadequate) market integration of renewable energies provides further measures that would create explicit demand for reserve capacity even without intervening in the market and politically mandating the reserve. For this to work effectively, renewable energies would need to bear a marketing responsibility and a balancing responsibility for their output. This would create a commercial incentive for them to secure their own back-up reserve if the energy is sold on a forward basis (in the way conventional generators already do today). They can achieve this through reserve capacity contracts with dispatchable generation from renewable or conventional technologies. With such reforms an energy-only market design may be selfsustaining.

4.6.2 Reforms with capacity mechanisms Where policy makers do not trust that the previously discussed reforms within the current energy-only regime are sufficient, they have options of employing capacity

78

Energy Market Design with Capacity Mechanisms

mechanisms as an additional insurance. This insurance should be designed such that it does not unnecessarily interfere with the functioning of the market. Even if this principle is respected, and the capacity mechanism is well designed, its implementation will nevertheless imply some additional (capacity) cost. Depending on how much trust policy makers place on the sustainable functioning of electricity markets we distinguish several scenarios: In a scenario where policy makers do not trust the sustainability of a reformed energy-only market, policy makers may opt for a comprehensive capacity mechanism covering all required capacity (eg capacity obligations or reliability options). In a scenario in which policy makers trust the (reformed) energy-only market in principle, but see a certain risk that their positive expectation may be disappointed or they see a temporary issue, policy makers require an insurance or a reserve that can be deployed if capacity adequacy is temporarily distorted or until a full comprehensive capacity mechanism can be implemented in case the energy-only market ultimately proves not to be sustainable. Provided that their concerns about the market not performing are small, a mechanism limited in scope is best able to address this. In case the perceived issue was more significant, then a more comprehensive capacity mechanism would be required. If the issue was perceived to be transitory then it would also be important to design the mechanism such that it would be reversible at a later stage. In a scenario in which policy makers have full trust in a (reformed) energy-only market no explicit capacity mechanisms would be required. In many countries the assessment of market failures and the costs and benefits of introducing capacity mechanisms to cure identified market is still pending. At any rate, policy makers should also consider that the introduction of capacity mechanisms comes with a risk of design flaws and policy failures. This could imply that the cost of the implementation of a capacity mechanism could exceed its benefit, especially if the market rules are poorly designed.

4.6.3 The European dimension has to be taken into account The European dimension adds a further layer of complexity. Coordination across countries is required regarding objectives and targets (how should security of supply be defined, and which target level should be achieved?), cross-border support and capacity exchange, and the interaction between capacity mechanisms in different countries. However, the need for cross-border participation raises complexity. At the minimum it should be ensured that capacity mechanisms are open to cross-border capacity. It should been ensured that there is sufficient interconnection capacity and that the contribution from abroad actually takes place in periods with scarcity. The Commission has addressed these issues, however, the willingness of Member States to coordinate across borders seems still to be at an early stage.

5 Different Approaches for Capacity Mechanisms in Europe Rationale and Potential for Coordination? Fabien Roques and Charles Verhaeghe

5.1 Introduction There is currently much debate about whether the lights will stay on in Europe without reform of the electricity market. In a number of Member States, the government and/or regulator have taken steps or are debating the introduction of capacity mechanisms in order to guarantee security of electricity supply. The paradox is that nowadays, many European countries are in a situation of overcapacity. Subsidized RES production displaces generation from thermal sources, which combined with the effect of the economic crisis on power demand, have dramatically reduced load factors for thermal plants. In addition, power prices have fallen to levels which do not reflect the whole generation cost. Instead, the current electricity prices reflect a temporary oversupply, as well as the downward pressure on prices associated with the development of renewables. There are very different drivers for the introduction of capacity mechanisms, including a range of alleged market failures in the currently energy-only markets.1 Irrespective of the economic debate on whether capacity mechanisms are needed, the key issue is that many market players and most policy makers believe that the current market and regulatory arrangements are unlikely to provide adequate investment incentives. Thus, most countries have taken steps to introduce capacity mechanisms or reform the existing ones, using very different approaches. The result is a patchwork of different national capacity mechanisms which may undermine the further integration of the European electricity markets. As a result, the Commission has taken steps to ensure a minimum level of coordination by explicitly including capacity mechanisms in its new EEAG 2014–2020.2 The objective of this chapter is to investigate the reasons for the differences in national approaches to capacity mechanisms, and to identify the potential for some degree of coordination across Member States. The chapter comprises five sections. In the following section (section 5.2) we focus on the specific local drivers for the implementation of capacity mechanisms in different Member States. Then, in section

1

For a more detailed discussion, see sections 1.1 and 4.2.3. Communication from the Commission, Guidelines on State aid for environmental protection and energy 2014–2020 [2014] OJ C 200/1–55 (EEAG 2014–2020). 2

80

Different Approaches for Capacity Mechanisms in Europe

5.3, we take a broader look at the integration of power markets in Europe and highlight potential market distortions which could arise from uncoordinated implementation of national capacity mechanisms. In section 5.4, we analyse the criteria of coordination set out in the new EEAG 2014–2020. The fifth and last section of this chapter (section 5.5) discusses the required steps necessary to enable coordination of capacity mechanisms across Member States.

5.2 A patchwork of capacity mechanisms in Europe: different designs fit different needs Most power markets in Europe were originally energy-only markets, which means that there was no specific mechanism to remunerate generators for their availability during peak hours or when the system was tight. In such markets, it is assumed that generation adequacy will be ensured because electricity prices will rise if market players anticipate an impending shortage of capacity, providing necessary investment incentives. This is grounded theoretically in the peak-load pricing theory, which shows that marginal pricing can provide fixed-cost recovery for investment based on the scarcity premium that all power producers earn when the system is tight.3 The underlying assumption is that power prices would climb to the level of VOLL4 at times of scarcity and that this would naturally lead market players to benefit from periods of high prices to remunerate their fixed costs.5 There is a rich academic literature focusing on the theoretical rationale for the implementation of capacity mechanisms. A range of market and regulatory failures are often put forward to justify the need for market intervention in the form of capacity mechanisms. For instance, regulatory interventions such as price caps often prevent prices from reaching remunerative levels at times of scarcity creating a revenue deficiency for generators. This, in turn, may cause insufficient investment in generation capacity (missing money problem, see section 1.1). The issue of the theoretical rationale for the introduction of capacity mechanisms is not the focus of this contribution, and readers interested in this issue are referred to the preceding chapters of this book,6 as well as further economic literature.7 3

See Figure 4.1 and accompanying text (section 4.2.1). See chapter 1, n 7 for explanation. 5 Marcel Boiteux, ‘La Tarification des demandes en point: application de la théorie de la vente au coût marginal’, Revue Générale de l’Electricité 58 (August 1949) pp 321–40. Marcel Boiteux, ‘La Tarification au coût marginal et les demandes aléatoires’, Cahiers du Séminaire d’Econometric 1 (1951) pp 56–9. Peter O. Steiner, ‘Peak loads and efficient pricing’ (1957) Quarterly Journal of Economics 71, 585–610. Hung-Po Chao, ‘Peak-load pricing and capacity planning with demand and supply uncertainty’ (1983) Bell Journal of Economics 14 (1), 170–90. Fred C. Schweppe, Michael Caramanis, Richard D. Tabors and Roger E. Bohn, Spot Pricing of Electricity (Kluwer Academic Publishers 1988). 6 In particular, chapters 1 and 4. 7 William W. Hogan, ‘On an “energy only” electricity market design for resource adequacy’ (Working paper, Center for Business and Government, Harvard University, 2005). Paul L. Joskow and Jean Tirole, ‘Reliability and competitive electricity markets’ (2007) The RAND Journal of Economics 38 (1), 60–84. Paul L. Joskow, ‘Capacity payments in imperfect electricity markets: need and design’ (2008) Utilities Policy 16 (3), 159–70. Peter Cramton and Steven Stoft, ‘Forward reliability markets: less risk, less market power, more efficiency’ (2008) Utilities Policy 16 (3), 194–201. Peter Cramton and Larry Ausubel, ‘Using forward markets to improve electricity market design’ (2010) Utilities Policy 18 (4), 195–200. 4

5.3 Toward coordination of capacity mechanisms in Europe

81

5.2.1 Different drivers for capacity mechanisms explain the range of approaches Faced with these concerns, many governments in Europe have implemented or are considering the introduction of capacity mechanisms. Figure 1.2 (section 1.2.3) provides an overview of different capacity mechanisms existing or being considered in Europe, reflecting a range of approaches. This includes strategic reserves, capacity obligations, capacity auctions, reliability options, and capacity payments.8 The patchwork of approaches across Europe originates in different country-specific drivers for market reform. First, countries have different security of supply concerns, ranging from generation adequacy issues, local imbalance issues with network constraints to problems with intermittent RES generation. Secondly, countries differ in the current and anticipated supply/demand balance, ie whether significant investment will be required or whether the key issue is to carry out an efficient rebalancing of an oversupplied system and manage stranded assets. Thirdly, the current and anticipated need for flexibility in the system depends on the country considered, depending on the level of interconnection with neighbouring countries, the share of flexible generation such as hydropower, or the existence of demand side response to manage the growth of intermittent generation. Finally, countries differ in their local market arrangements and whether special design features such as price caps and/or constraints to scarcity pricing lead to a shortfall of revenues for existing units. Table 5.1 summarizes the local issues related to security of supply for the five largest European countries and highlights the very different drivers of capacity mechanisms implementation. In some countries (for instance, France), the key security of supply concern is to meet the anticipated increase in peak demand.9 In other countries such as Germany, security of supply concerns originate in regional imbalances and network constraints.10 In the UK, which is weakly interconnected and faces massive thermal plants retirements, the key issue is to drive new investment,11 whilst countries in Southern Europe such as Italy and Spain face a lasting oversupply, and are confronted with the issue of managing a smooth rebalancing of the system with stranded assets.12 Looking forward, capacity mechanisms which have originally been designed to address generation adequacy issues and missing money will likely need to evolve and take into consideration the growing impact of intermittent renewables on power market dynamics.

5.3 Toward coordination of capacity mechanisms in Europe The patchwork of national approaches to capacity mechanisms raises questions about the lack of coordination between Member States. As the previous section 5.2 highlighted the different drivers of capacity mechanisms across Europe, it is unlikely that a common approach at the European level will be practical or even suitable. However, there would 8 9 11

See section 1.2.3.4 for a description of reliability options, and section 1.2.3.5 for capacity payments. 10 See chapter 14 on France. Chapter 15 on Germany. 12 Chapter 22 on the UK. Chapters 17 (Italy) and 21 (Spain).

Table 5.1 Local drivers of the introduction of capacity mechanisms France Local specificities – Thermo sensitivity of power demand (electric heating) – Peak demand growth

Key issues

Main objectives of capacity mechanisms

– Peak demand growth (+25% in 10 years) – Missing money for peak plants – Low profitability of CCGTs

Germany

UK

Spain

Italy

– Grid constraints from North to South – Nuclear phase-out – Strong RES growth

– Large retirements of thermal plants – Limited interconnection – Strong RES growth

– Weak demand – Strong RES growth – Limited interconnection – Quasi-mandatory13 pool

– Capacity needs in Southern Germany – Flexibility needs

– Major investment needs (capacity gap) – Retirements driven by Large Combustion Plant Directive15 and Industrial Emissions Directive16 – Need for flexibility

– Overcapacity and low – Overcapacity and low profitability of CCGTs profitability of CCGTs – Generation back-up – Coordination of necessary due to RES generation and penetration network investment

– Low profitability of CCGTs

– Internal zones and grid constraints – Strong RES growth – Central dispatch14

– Flexibility needs

– Ensure generation – Retain existing capacity in the – Ensure generation adequacy – Incentivize availability – Incentivize availability adequacy Southern Germany & drive and flexibility of and flexibility of – Support the new investment – Drive new investment in CCGTs existing plants existing plants development of – Ensure availability of flexible – Manage smooth – Manage smooth demand response back-up generation rebalancing / avoid rebalancing / avoid – Prevent market power – Ensure availability of flexible back- massive retirements massive retirements abuses up generation – Limit price spikes & –Prevent market power volatility abuses

Source: Authors’ own analysis.

13 All the available generation units are obliged to submit offers in the day-ahead auction for their production not subject to bilateral contracts. On the other hand, the production covered by bilateral contracts has to be offered by each generation company in the day-ahead auction at a price reflecting its opportunity costs (regulatory obligation for the optimization of the generation portfolios). 14 Central dispatch, as opposed to self-dispatch, is an arrangement where the unit commitment scheduling is centralized and market participants are given their position based on a central decision. The TSO determines the dispatch values and issues instructions directly to generators (or demand). Conversely, in the self-dispatch model generators determine a desired schedule and commitment for themselves based on their own economic criteria. 15 Directive 2001/80/EC of the European Parliament and of the Council of 23 October 2001 on the limitation of emissions of certain pollutants into the air from large combustion plants [2001] OJ L 309/1. 16 Directive 2010/75/EU of the European Parliament and of the Council of 24 November 2010 on industrial emissions (integrated pollution prevention and control) [2010] L 334/17.

5.3 Toward coordination of capacity mechanisms in Europe

83

be merits in working toward some degree of coordination in order to minimize the potential distortions associated with different capacity mechanism designs.

5.3.1 Capacity mechanisms in the wider context of EU power markets integration There are concerns about implementing different and potentially incompatible capacity mechanisms, particularly because power markets in Europe are strongly interconnected. The risk is that uncoordinated national measures could undermine further integration of European power markets and create distortions in electricity markets. Some of the key differences in the design of capacity mechanisms stem from local specificities of electricity systems across Member States. In this context the coordination and possible harmonization of capacity mechanisms should be considered a part of the wider issue of defining a common ‘Target Model’ for electricity markets in Europe. The EU Target Model is a set of proposals for the creation of an internal market for energy in Europe. It has been elaborated, mainly for cross-border power trade, through a long consultation and regulatory process. The EU Target Model is based on the principle of market coupling, a method to integrate national wholesale electricity markets,17 and has ultimately materialized in a draft Regulation establishing a Guideline on Capacity Allocation and Congestion Management (draft CACM)18 adopted by Member States through comitology on 5 December 2014, resulting from the Network Code process. The definitive adoption of CACM is expected in early 2015, after the scrutiny from the European Parliament and Council. Indeed, the Third Energy Package19 adopted in July 2009 marked a significant step in the integration of European power markets. It defined a process for harmonizing some of the key building blocks of power markets, through the drafting of Network Codes and the Framework Guidelines. Network codes are a set of rules drafted by ENTSO-E, with guidance from ACER, to facilitate the harmonization, integration, and efficiency of the European electricity market. ENTSO-E was mandated by the Commission to draft these rules for electricity following the principles set up by ACER in the Framework Guidelines. Under development since 2011, each code takes approximately eighteen months to complete. Following ACER’s recommendation, each code is submitted to the Commission for approval through the comitology process, to then be voted into EU regulation, in order to be directly—ie without transposition— implemented across Member States. This process is illustrated in Figure 5.1. 17 According to the Commission, the EU Target Model includes the day-ahead market coupling, the intraday, balancing, and cross-border forward markets. See the Commission’s, Public consultation on the governance framework for the European day-ahead market coupling (D(2011) 1176339, 28 November 2011, https://ec. europa.eu/energy/sites/ener/files/documents/20120229_market_coupling.pdf, accessed 1 February 2015, p 1. 18 Draft Regulation establishing a Guideline on Capacity Allocation and Congestion Management, provisional final version of 5 December 2014 available at http://ec.europa.eu/energy/sites/ener/files/docu ments/cacm_final_provisional.pdf, accessed 1 February 2015. 19 See, in particular, Directive 2009/72/EC of the European Parliament and of the Council of 13 July 2009 concerning common rules for the internal market in electricity and repealing Directive 2003/54/EC [2009] OJ L 211/55 (2009 Electricity Directive) and Regulation (EC) 714/2009 of the European Parliament and of the Council of 13 July 2009 on conditions for access to the network for cross-border exchanges in electricity and repealing Regulation (EC) 1228/2003 [2009] OJ L 211/15 (2009 Cross-border Regulation).

84

Different Approaches for Capacity Mechanisms in Europe Priority areas defined by the European Commission EC invites ACER to develop FG EC reviews FG and invites ENTSO to develop NC ENTSO submits the NC to ACER for evaluation ACER provides an opinion and recommends NC to EC for adoption

Entso Scoping

FG

NC

6 months

12 months

EV

Comitology

3m

Figure 5.1 Network Code process Source: ACER. NC—Network Codes, EC—European Commission, FG—Framework Guidelines, EV—Evaluation.

The progress for the implementation of the Target Model, and in particular some of the Framework Guidelines and Network Codes, is facing a number of hurdles. Indeed, the process to draft Network Codes takes longer than expected, both during the drafting phase amongst TSOs and in interaction with ACER and national regulatory authorities (NRAs), and during the comitology phase to validate and publish the Network Codes, thus transformed into a Regulation. Furthermore, the implementation takes time and the political objective to achieve the internal electricity market by 201420 is hardly achievable in practice. Despite some progress in market coupling, many ongoing projects are unlikely to be implemented by the end of 2014 (eg intraday markets or flow-based cross-border capacity calculation) and deadlines to improve balancing markets go beyond 2020. Table 5.2 highlights the key different approaches for the different building blocks of electricity markets across Europe. Historically in Europe, capacity mechanisms were introduced in countries with limited interconnection capacity and/or internal grid constraints, which opted for price-based mechanisms (regulated capacity payments). These countries, such as Ireland, Spain, Italy, or Greece, also implemented a centralized dispatch system in order to have more control on congestion management.21 Debates on the European Target Model have broadly converged to day-ahead auctions through market coupling and continuous cross-border intraday trading until close to real time; although self-dispatch was not imposed, this orientation challenges the design of central dispatch models. Furthermore, capacity mechanisms’ design seems to move towards market-based designs, with the implementation of capacity obligations in France and capacity auctions in the UK and the ongoing reforms in Italy, Ireland, or Greece.22 20 European Council, Conclusions (EUCO 75/1/13 REV 1, 22 May 2013) in particular point 2, available at http://www.consilium.europa.eu/ueDocs/cms_Data/docs/pressData/en/ec/137197.pdf, accessed 1 February 2015. See chapter 1, text accompanying n 3. 21 See chapters 21 on Spain, 17 on Italy, and 16 on Greece. 22 See chapters 14 on France, 22 on the UK, 17 on Italy, and 16 on Greece.

5.3 Toward coordination of capacity mechanisms in Europe

85

Table 5.2 Integration of EU power markets: The different building blocks

Forward market

Ireland

Spain, Portugal, Italy

Nordic, CWE regions

Great Britain

– No meaningful forward market

– Financial forward market

– Financial and physical forward markets

– Mainly physical forward market

– Quasimandatory dayahead auction – Locational bidding

– Day-ahead auction – Portfolio bidding

– No particular significance of DA – Portfolio bidding

– D-1 gate closure – Intraday auction – Continuous trading – No intraday market slots – H-4 gate closure – H-1 gate closure

– Continuous trading – H-1 gate closure

Day-Ahead – Central dispatch market – Traded volumes/ prices not firm – Locational bidding Intraday market

Capacity – Fixed capacity mechanism payment

– Capacity and availability payment

– Strategic reserve (Nordic – Centralized region, Belgium, forward capacity Germany) market – Decentralized capacity market (Fr)

Source: Authors’ own analysis. CWE—Central and Western Europe, DA—Day Ahead, D—Day, H—Hour.

The different approaches presented in Table 5.2 raise the question as to how much harmonization of the key market design building blocks—including capacity mechanisms—is required as a prerequisite to further market integration. We suggest that capacity mechanisms should be an integral part of the Framework Guidelines and Network Codes process, in order to drive some coordination in a consistent way between the different building blocks. In other words, since the issues of power market design and capacity mechanisms are so closely interlinked, an integrated approach is necessary and the Network Code process could be extended to include capacity mechanisms. However, the difficulty to identify a common approach toward security of supply and capacity mechanisms at the European level suggests that a regional approach might be the best way forward to drive some coordination. Indeed, many of the key drivers for the introduction of capacity mechanisms are similar on a regional basis, and the creation of the Regional Initiatives (RIs) have proven to be an efficient bottom-up alternative for market integration compared to the more EU-driven top-down processes.23

5.3.2 Potential distortions associated with uncoordinated capacity mechanisms There are concerns that implementing different capacity mechanism designs in countries that border one another could introduce distortions in the integrated electricity markets. Having different neighbouring capacity mechanisms could, in the short term, affect the operational efficiency of plant dispatch, and, in the long term, undermine the dynamic efficiency of power markets by skewing generation and transmission investment incentives. 23

See eg, Everis and Mercados, From regional markets to a single European market, Report 2010.

86

Different Approaches for Capacity Mechanisms in Europe

In the short term, capacity mechanisms can affect price formation in electricity markets. Depending on their design, they can affect the mark-up over short run marginal cost (SRMC) that allows fixed cost recovery. As a consequence, if one of the two interconnected countries does not have a capacity mechanism, or if their capacity mechanisms significantly differ in design the mark-up over SRMC will be different across the two borders for plants clearing the market. This could lead to distortions in the dispatch of the marginal plants across the border, where a plant with higher SRMC and a lower mark-up would be dispatched before a plant with a lower SRMC and a higher mark up. This could affect interconnection flows, and undermine the short-term dispatch efficiency of coupled markets, resulting in a loss in welfare. In the long term, the altered power price dynamic, as well as different complementary capacity remuneration in the neighbouring countries will also affect investment incentives, and more precisely the location of new plants across borders. The crosssubsidization effect of supporting the existing and new capacity through a capacity mechanism in one country in order to export energy to another country is often referred to as ‘capacity leakage’, studied by Creti and Fabra,24 as well as Finon.25

5.4 Capacity mechanisms and the new EEAG 2014–2020: Key issues In order to ensure that national capacity mechanisms do not unduly distort the internal market, the Commission has recently taken on the task of identifying best practices in their design, by including them in the new EEAG 2014–2020.26 Based on the premise that capacity mechanisms may constitute state aid, the EEAG set out detailed conditions for their approval. In the following we discuss some of the key issues likely to arise in the implementation of the EEAG. As our discussion centres on the points of the EEAG which are particularly relevant from the economic perspective and in the context of this chapter, we refer readers to chapters 1 and 9 for a more comprehensive legal and policy analysis of the issue.

5.4.1 Should state aid rules apply to capacity mechanisms in the first place? Generally speaking, state aid can occur whenever state resources are used to provide support that gives certain undertakings an economic advantage over others, distorting competition and cross-border trade. To determine whether a capacity mechanism is captured by the notion of state aid, the characteristics of that particular mechanism would need to be considered. Only if a mechanism combines all the elements of the state aid definition, can it be considered state aid.27 24 Anna Creti and Natalia Fabra, ‘Supply security and short-run capacity markets for electricity’ (2007) Energy Economics 29 (2), 259–76. 25 Dominique Finon, ‘Can we reconcile different capacity adequacy policies with an integrated electricity market?’ Economics Papers from University Paris Dauphine (2013). 26 EEAG 2014–2020 (n 2). 27 For a comprehensive discussion, see chapter 9.

5.4 Capacity mechanisms and the new EEAG 2014–2020

87

This exercise might turn out to be challenging (if at all possible), depending on the type of capacity mechanism in question. For instance, where compensation provided under a capacity mechanism takes the form of direct payments financed from the state budget (capacity payments),28 some elements of the state aid definition, like economic advantage and the link with state resources, are readily identifiable. Other types of capacity mechanisms might be less straightforward. In the case of capacity obligations, for example, the burden falls on suppliers to certify that they have at their disposal enough capacity to supply their customers.29 Not only is it much more difficult to translate this obligation into an economic advantage on the generator’s part, but it is also less clear who bears the corresponding cost of this obligation. It may be passed on to final consumers, but it may also remain at the supplier’s level if the competition between suppliers is strong enough. Given the difficulties in assessing such design types, one may question the application of state aid rules to capacity mechanisms in the first place. In any case, the EEAG 2014–2020 set out the Commission’s criteria in assessing the compatibility of capacity mechanisms with state aid rules and we discuss them in the next sections.

5.4.2 Does the capacity mechanism aim at a well-defined objective of common interest? First, the capacity mechanism’s objective shall be clearly identified.30 More precisely, the problem that the mechanism aims at solving must be carefully analysed and its identification must be consistent with the ENTSO-E generation adequacy assessment, taking into account the contribution of cross-border trade and interconnectors. The causes of the problem must also be studied to understand why the market would not deliver the necessary capacity without public intervention. Regulatory and market failures must be analysed. Indeed, the existence of distortive market arrangements, such as price caps, barriers to the development of DSR, or ill-designed RES support schemes may account for generation adequacy problems. In the case where such regulatory barriers are identified, they should a priori be addressed to the extent possible, before deciding on implementing a capacity mechanism. In addition, developing further interconnection capacities or measures to encourage DSR might be considered a more efficient solution to the identified problem, at least as transitory measures by the time all the regulatory barriers are removed.

5.4.3 Is the aid well designed to address the problem? Once the market or regulatory failure has been clearly defined, an appropriate and proportionate response should be designed.31 In particular, the capacity mechanism

28 29 30

For more detail on capacity payments, see section 1.2.3.5. For more detail on capacity obligations, see section 1.2.3.3. 31 EEAG 2014–2020 (n 2) paras 220–2. EEAG 2014–2020 (n 2) paras 222–32.

88

Different Approaches for Capacity Mechanisms in Europe

should provide incentives for operators to contribute to solving the adequacy problems. A mechanism which does not incentivize operators to adapt their actions to eliminate the problem is likely to be discarded, notably because it would likely create windfall profits. In order to avoid distortions, the Commission considers that capacity mechanisms should remunerate the commitment (and not the sale of energy) to deliver capacity when required and provide incentives to both existing and planned generating units, and to any operators which could contribute to solving the problem, including demand side response and storage solutions. Member States should also consider to what extent interconnection capacity could remedy the problem. In addition, the mechanism should not lead to unreasonable rates of return and windfall profits, and should not discriminate between existing and future generation. A transparent and nondiscriminatory open competitive bidding process would be a good solution to address this point.

5.4.4 Are the distortions of competition and the effect on trade limited or avoided, so that the overall balance is positive? The capacity mechanism should avoid negative effects on the market and should not constitute a barrier to market integration. For instance, export restrictions32 or wholesale price caps should be avoided. The measure should have a positive impact on competition, or at least it should not unduly strengthen dominant positions. If technically possible, the mechanism should allow the participation of operators located in other Member States. Finally, the mechanism should not go against other EU policies. In particular, it should not reward investments in fossil fuel power plants, unless there are no alternative or environmentally less harmful measures. On the contrary, it should give preference to low carbon generators in case of equivalent technical and economic parameters. It should also be consistent with RES policy objectives and support schemes. In sum, the EEAG 2014–2020 impose boundaries on the design of national capacity mechanisms in order to limit their potential negative impact on the internal market and enable cross-border participation. However, in our view, the EEAG are insufficiently prescriptive to drive harmonization and reinforce coordination towards a more pan-European solution, probably for good reasons, as so far no one-size-fits-all approach has been identified nor seen as credible. The next section therefore explores in greater detail the potential approaches for coordination of national capacity mechanisms and opening them to capacity located abroad.

32 For example, the capacity mechanism recently implemented in Russia gives Russian generators incentives not to export during peak hours despite Finnish prices being higher than Russian prices. See for instance the report ‘Cross-border electricity trade between the Nordic “energy-only” market and the Russian capacity-based market’, Lappeenranta University of Technology, 2013.

5.5 Coordinating capacity mechanisms in Europe

89

5.5 Coordinating capacity mechanisms in Europe 5.5.1 Where to start to coordinate capacity mechanisms? A primer While conceiving a common capacity mechanism design for Europe is unlikely for the reasons set out in the previous sections, there is merit to driving some coordination in design in order to minimize potential market distortions. In the following we identify a number of preliminary steps which we consider prerequisites for coordination of national capacity mechanisms. The first crucial step consists in defining explicit criteria to measure the level of supply reliability in each country and ensuring their consistency. Certain countries, for instance France or the UK, adopted the so-called LOLE criterion. LOLE stands for the loss-of-load expectation and represents the number of hours (or days) per year in which, over the long term, it is statistically expected that supply will not meet demand. Other countries, like Germany or Spain, adopted capacity margin criterion, according to which installed capacity needs to exceed peak demand by a certain margin deemed to be sufficient to provide a reasonable level of supply security. Finally, there are also countries which have no explicit supply security criteria, but rather rely on engineering principles to evaluate the necessary investments to upgrade or reinforce networks. This is the case for the Netherlands, Switzerland, and the Nordic countries. Interestingly, the different security of supply standards across Europe imply that the problem of capacity leakage (cross-subsidization)33 predates the implementation of capacity mechanisms, in the sense that countries with stricter security of supply standards actually crosssubsidize countries with lower security of supply standards. In our view, coordination of capacity mechanisms requires prior harmonization of the different supply security criteria in Europe. In that respect, the European TSOs, through their umbrella association ENTSO-E,34 have developed best practices in terms of forward generation adequacy assessments. ENTSO-E’s Ten-Year Network Development Plan (TYNDP)35 released every two years shows some convergence in assessment methods, but points towards the need to define a common methodological framework in that respect. The second important step to enable coordination of capacity mechanisms is defining common certification and verification procedures for generation units and demand-side resources that will participate in capacity mechanisms across borders. National TSOs would need to work together in that respect in order to, at the very least, establish a common registry of generation and demand-side resources, as well as to develop a common approach to certify and verify their availability in line with the definition of the capacity product. Last but not least, coordination of capacity mechanisms requires that TSOs develop a common framework to deal with situations of system stress at a regional

33

See Creti and Fabra (n 24) and Finon (n 25). ENTSO-E stands for the European Network of Transmission System Operators for Electricity, an association of Europe’s transmission system operators (TSOs) for electricity. 35 The Ten-Year Network Development Plan (TYNDP) 2014 is available at https://www.entsoe.eu/ major-projects/ten-year-network-development-plan/tyndp-2014/Pages/default.aspx, accessed 1 February 2015. 34

90

Different Approaches for Capacity Mechanisms in Europe

level, including the harmonization of operational rules. Capacity shortage in one or two countries would require clear rules and corresponding operational practices in place to ensure the physical delivery of energy which has been contracted for. All these preliminary steps require close collaboration of TSOs and regulators, and a practical way forward would be to set up regional task forces. Whilst the EU-wide harmonization led by ENTSO-E should continue, regional approaches have proven to be a successful way to find pragmatic solutions, and national TSOs have a long history of cross-border cooperation.

5.5.2 Why take into account cross-border contribution? In the internal electricity market, both domestic and external generators contribute to security of supply. For instance, during the European cold wave of February 2012, demand for electricity in France hit a new record of up to 102 GW and load shedding would have been necessary without interconnectors’ contribution to meeting peak demand. The increasing integration of European markets reinforces interdependencies between countries. In this context, the Commission is concerned that implementing a capacity mechanism with no cross-border participation would lead to inefficiencies and distortions to the detriment of the internal market. In particular, the Commission fears that such a capacity mechanism would distort flows on the interconnector and would steer new investments away from neighbouring markets by undermining the financial viability of generation located there. This would not only increase the missing money problem and jeopardize generation adequacy and security of supply in the neighbouring countries, but also, due to the previously mentioned interdependencies, in the whole region.36 Therefore, the EEAG 2014–2020 attach great importance to cross-border aspects of capacity mechanisms. Member States shall respect their commitments to export even during periods of high demand in their country.37 This is in line with the 2009 CrossBorder Regulation according to which ‘transaction curtailment procedures shall only be used in emergency situations where the transmission system operator must act in an expeditious manner’,38 and also with the Security of Supply Directive which provides that Member States shall not discriminate between national and cross-border energy contracts in situations of system stress.39

5.5.3 How to open capacity mechanisms to cross-border capacity? There are several ways (or degrees) of taking interconnectors and imports into account during scarcity periods. (a) First, Member States could simply consider the statistical 36 Commission staff working document, Generation Adequacy in the internal electricity market— guidance on public interventions, 5 November 2013 (Generation Adequacy SWD) pp 28–30. 37 Generation Adequacy SWD (n 36) pp 28–30. 38 2009 Cross-border Regulation (n 19) Art 16(2). 39 Directive 2005/89/EC of the European Parliament and of the Council of 18 January 2006 concerning measures to safeguard security of electricity supply and infrastructure investment [2006] OJ L 33/22 (Security of Supply Directive) Art 4(3).

5.5 Coordinating capacity mechanisms in Europe

91

contribution of imports when assessing their level of generation adequacy (implicit cross-border participation).40 (b) Secondly, Member States could implement different capacity mechanisms, but allow for cross-border trade in capacity (explicit crossborder participation).41 (c) Finally, a more radical approach would be for the Commission to identify a ‘reference model’ for capacity mechanisms and force some harmonization across Member States on a regional, and/or European basis. As the reasons for implementing capacity mechanisms vary across Europe, and several Member States already have (or are implementing) capacity mechanisms, it is very unlikely to envisage a European capacity mechanism in the short/medium term, or even to expect a strong harmonization and coordination between different national models, with the exception of a few cases of closely integrated regional markets (such as Spain and Portugal, or in the Nordic region). As a consequence, in the short term, the Commission’s focus is mainly on whether and how to include interconnectors and foreign capacity in national capacity mechanisms (options (a) to (c) noted earlier). According to the EEAG, ‘the measure should [ . . . ] take into account to what extent interconnection capacity could remedy any possible problem of generation adequacy.’42 This would indicate that, at the very least, Member States have to take cross-border resources into account when assessing generation adequacy. However, the wording of the EEAG leaves a certain margin of discretion as to how this can be done in practice. Namely, it is stated that ‘the measure should be designed [ . . . ] taking into account [ . . . ] the participation [ . . . ] of operators offering measures with equivalent technical performance, for example [ . . . ] interconnectors.’43 This could advocate direct participation of interconnectors in a capacity mechanism. It is also mentioned in the EEAG 2014–2020 that capacity mechanisms should consider ‘the participation of operators from other Member States where such participation is physically possible in particular in the regional context, that is to say where the capacity can be physically provided to the Member State implementing the measure and the obligations set out in the measure can be enforced.’44 This suggests that the Commission favours, where feasible, direct participation of foreign operators in a capacity mechanism. Table 5.3 compares three ways in which cross-border capacity can participate in a capacity mechanism. Several key elements differentiate these different approaches. First, the amount of capacity that will be procured is a priori lower in the first option as the cross-border contribution is netted off the capacity requirement. Second, the cost and revenue sharing differ. In the first option, neither interconnectors nor generators receive capacity revenues: the end consumers benefit from their contribution ‘free of charge’. In the second option, the interconnector directly participates in the capacity mechanism and receives the full capacity price. In the third option, foreign generators directly participate in the capacity mechanism and receive the capacity price, but revenues may be shared with interconnectors through the mechanism to access 40 42 44

See section 3.4.1 and chapter 6. EEAG 2014–2020 (n 2) para 232 (a). EEAG 2014–2020 (n 2) para 232 (b).

41

See section 3.4.1 and chapter 6. 43 EEAG 2014–2020 (n 2) para 232 (a).

92

Different Approaches for Capacity Mechanisms in Europe

Table 5.3 Cross-border participation in a capacity mechanism: Possible models No participation by interconnectors

Key features

Statistical contribution of interconnectors in the assessment of generation adequacy as a reduction of the overall volume required to be procured No payment to interconnectors

Participation by interconnectors

Interconnector participates in capacity mechanism Interconnector acts as intermediary between capacity mechanism and foreign generation

Participation by foreign generation

Foreign generation can pariticipate provided that they can demonstrate sufficient access to interconnectors

Very complex implementation Complex implementation Easiest implementation

Assessment

Undermines dynamic efficiency (underinvestment in interconnection)

Strong investment incentives as interconnection captures full value of capacity mechanism Key issue lies in ability of interconnector operator to control power flow

Investment incentives in interconnection depend on how capacity revenues are shared between interconnectors and generators Need for mechanism to allocate interconnection capacity

Source: Adapted from Fabien Roques’ presentation at the EPRG Conference (May 2014).

interconnectors for the capacity mechanism’s participation. Another difference between the models is that—unless cross-border capacity is unlikely to be congested—it will probably be necessary to implement a mechanism to allocate access to interconnectors’ capacity. Indeed, as interconnectors’ capacity is limited, it should be allocated in an efficient and non-discriminatory manner to foreign generators willing to participate in a neighbouring capacity mechanism (see section 5.5.4).

5.5.4 What are the prerequisites for cross-border exchange of capacity? If country A implements a capacity mechanism with cross-border exchange of capacity, capacity located in country B may bid in the capacity mechanism of country A, but only up to the available transmission capacity, as illustrated in Figure 5.2. However, in practice, exchanging capacity between countries with different capacity mechanisms (or in the case where only one of the countries has a capacity mechanism in place) raises several implementation issues, such as how to certify the foreign capacity, how to deal with cross-border transmission capacity, how to make sure the foreign capacity is comparable to internal capacity and provide the same product/ service to security of supply or what would happen in case of scarcity in both countries. In its November 2013 Communication,45 the Commission suggests a few approaches to deal with some of these implementation issues. However, the Commission acknowledges 45 Communication from the Commission, Delivering the internal electricity market and making the most of public intervention, C(2013) 7243 final (November 2013 Communication).

5.6 Conclusions

93

Exchange of capacity

Peak-load

Capacity Transmission capacity

Country A

Capacity Peak-load

Country B

Figure 5.2 Cross-border exchange of capacity Source: FTI-CL Energy.

that there may be practical difficulties in implementing a framework for explicit crossborder participation and considers, as an interim solution, the implicit cross-border participation (ie taking into account the statistical contribution of imports in the assessment of generation adequacy).46 For instance, certifying capacity located abroad implies that either the national certification body (usually the TSO) would receive access to all the relevant data, for instance related to the availability and the technical performance of the generator located outside its control area, or that it would need to cooperate with its counterparty of the neighbouring country, the mission and powers of which may be different. If no capacity mechanism is implemented in the neighbouring country, there might even be no legal framework in place which would allow for the certification process and the controls stemming from the home country. Another problem is to define the preconditions to ensure that internal and cross-border capacities may participate on an equal footing in the capacity mechanism, taking into account the access to interconnection and the rules applicable in the energy, balancing, and capacity markets.

5.6 Conclusions Capacity mechanisms are currently introduced or reformed in different EU Member States because many market players and most policy makers believe that the current market and regulatory arrangements are unlikely to provide adequate investment incentives. This results in a patchwork of approaches and raises concerns about the potential impact of uncoordinated capacity mechanisms on the integration of European energy markets. This chapter analyses the local drivers for the implementation of capacity mechanisms in order to understand the feasibility of a common or coordinated approach for capacity mechanisms, either in terms of cross-border participation or harmonization, up to an EU-wide mechanism. Our analysis shows that the drivers vary significantly. They depend on the local electricity system needs and on whether the key issue is one of investment need, of local network constraints, or of intermittency management. 46

Generation Adequacy SWD (n 36) pp 28–30.

94

Different Approaches for Capacity Mechanisms in Europe

They are also impacted by the local electricity market arrangements and whether special design features such as price caps and/or constraints to scarcity pricing lead to a shortfall of revenues for existing units. Due to these differences, an EU-wide approach, ie a harmonized capacity mechanism covering the whole EU, is unlikely, and effort should focus on ensuring a minimum level of coordination across neighbouring countries to minimize potential distortions. Moreover, given that the drivers for the implementation of capacity mechanisms are often shared on a regional basis, a regional approach towards capacity mechanisms would make the most sense. Furthermore, we discuss the Commission’s approach in assessing capacity mechanisms under the state aid rules. The Commission’s EEAG 2014–2020 constitute a first step towards more coordination of capacity mechanisms. However, they remain insufficient to foster a strong coordination of national capacity mechanisms or to implement regional capacity mechanisms. Coordination cannot be achieved without a minimum level of harmonization and agreement on the objectives and principles of capacity mechanisms. In the next step, we formulate a number of preliminary steps which we consider prerequisite for the coordination of national capacity mechanisms. First, we identify the need to define explicit reliability criteria and to use a common methodological framework for generation adequacy assessment. Then, we recommend harmonizing TSOs’ operational practices in order to align certification and verification procedures for power plants and demand-side resources across the EU, or at least at the regional level. Moreover, it will likely be necessary to develop a cooperation framework, including operational rules, to deal with situations of system stress. Finally, we focus on the issue of cross-border participation in capacity mechanisms. We identify three approaches and highlight possible difficulties in their implementation. We find that despite these practical problems, developing cross-border exchange of capacity might force Member States to address some key issues of market integration and harmonization, ultimately leading to regional solutions or even to a pan-European approach.

6 Capacity Mechanisms and Cross-Border Participation The EU Integrated Approach in Question Dominique Finon1 The implementation of capacity mechanisms by Member States wishing to ensure generation adequacy is under the strict scrutiny of the Commission. Beyond the discussion around the compliance of these national capacity mechanisms with EU energy law2 and state aid control,3 an important part of this debate focuses on explicit cross-border participation with bilateral exchanges of capacity rights from external generators to the capacity adequacy system of a neighbouring country managed through a capacity mechanism. This chapter presents, in section 6.1, common rationale for explicit cross-border participation in legal and economic terms. In section 6.2 it discusses the physical and economic relevance of explicit cross-border participation in comparison to the simpler implicit cross-border participation, consisting in consideration of the statistical contribution of an external system to the reliability of its neighbouring system. Sections 6.3 and 6.4 examine the social benefits of trade in capacity rights in explicit cross-border participation from the point of view of one system, and then the EU level of multiple systems.

6.1 Elements of the EU debate on cross-border participation In EU policy terms, cross-border participation in capacity mechanisms is often considered as the solution most consistent with the increasing integration of the wholesale electricity markets.4 In that respect, the Commission’s EEAG 2014–2020 consider that capacity mechanisms ‘should be designed in a way so as to make it

The author wishes to thank Charles Verhaeghe for his comments and advice on the first version of this chapter. Thanks are also due to Chaire European Electricity Markets of the ParisDauphine Foundation, supported by RTE, EDF, EPEX Spot, and the UFE. The views and opinions expressed in this chapter are those of the author and do not necessarily reflect those of the partners of the CEEM. 2 In particular, Directive 2009/72/EC of the European Parliament and of the Council of 13 July 2009 concerning common rules for the internal market in electricity and repealing Directive 2003/54/EC [2009] OJ L 211/55 (2009 Electricity Directive), in terms of public service obligations. 3 For a detailed discussion, see chapter 9. 4 In all time frames from multi-year forward to real time (forward, day-ahead, intraday, balancing) on energy and transmission rights. 1

96

Capacity Mechanisms and Cross-Border Participation

possible for any capacity which can effectively contribute to addressing the generation adequacy problem to participate in the measure, in particular, taking into account [ . . . ] the participation of operators from other Member States where such participation is physically possible’.5 In the same vein, the Generation Adequacy SWD states that ‘a mechanism which excludes cross-border participants could result in new generation capacity displacing imports. This would undermine the financial viability of generation in other Member States and could have a negative impact on regional security of supply’.6 This position, on the cross-border trade of capacity products, is backed by Article 4(3) of the 2006 Security of Supply Directive, which refers to the Treaty’s provisions on the free movement of goods within the EU and explicitly forbids discrimination by Member States between cross-border contracts and national contracts.7 The economic rationale of the Commission’s approach is that, in the internal electricity market, both domestic and cross-border capacities contribute to delivering two public policies objectives: supply reliability (ie the shortterm security of supply) and capacity adequacy (ie the long-term guarantee of supply reliability). In the terms of public economics theory, supply reliability and capacity adequacy are collective goods. Although specific to each national system, they are in fact so interdependent with those of the other systems that they constitute collective goods which are common to all these systems. But this applies, when considering two systems, with the caveat that they are only common to both systems if there is no congestion on the interconnection from one system to the other when the latter is in critical period. This interdependency will further increase with EU market integration. Accordingly, these multi-systems collective goods should ideally be managed at an integrated EUwide market level within a common capacity mechanism, with eventual bidding zones to take into account the physical reality of congested interconnections during critical periods. If it is not possible to implement a common capacity mechanism, then the next best alternative is to have individual capacity mechanisms, but of similar design to avoid distortive effects between respective energy markets and capacity mechanisms, and open to capacity located in other Member States. This can be defined, in other words, as a system of harmonized capacity mechanisms which allows an ideal crossborder exchange of capacity products.8 However, if it is not possible to harmonize capacity mechanisms, the cross-border exchange of capacity products is still a socially efficient option,9 provided that foreign generators can offer the exact same capacity

5 Communication from the Commission, Guidelines on State aid for environmental protection and energy 2014–2020 [2014] OJ C 200/1–55 (EEAG 2014–2020) para 232. 6 Commission staff working document, Generation Adequacy in the internal electricity market— guidance on public interventions, 5 November 2013 (Generation Adequacy SWD) p 28. 7 Directive 2005/89/EC of the European Parliament and of the Council of 18 January 2006 concerning measures to safeguard security of electricity supply and infrastructure investment [2006] OJ L 33/22 (Security of Supply Directive) Art 4(3). 8 For the concept of property rights on reliability and capacity adequacy see section 6.1.1. 9 Despite the heterogeneity of different capacity products emanating from the difference of capacity mechanisms.

6.1 Elements of EU debate

97

products as domestic producers, while also allowing the option of opting out of the capacity mechanism of their home system, should it exist there, to make offers in the neighbouring capacity mechanisms. This EU approach to supply reliability and capacity adequacy can be interpreted in the following way. System X is required to rely on any generation plant capacity located in system Y that chooses to forward commit to provide a guarantee of energy supply in system X’s capacity mechanism during critical periods of the latter. However, this reliance on bilateral capacity contracts becomes risky if the TSO of Y has the possibility to revoke the energy forward committed by generators of Y in capacity contracts with X during system Y’s critical periods. So, as said, Article 4(3) of the Security of Supply Directive forbids discrimination between cross-border transactions and internal transactions. However, when both systems have the same critical period, the contracted capacities in system Y cannot be replaced by anything else for the supply reliability of Y, and cross-border trade (and the reliability rights emanating from this trade) become problematic. In other words, cross-border capacity exchange compels each TSO to give up some control over its own system’s capacity adequacy, as it has to share all the reliability rights of its own system with those of the other systems in the intraday and real-time frames. This mutual sharing takes place even if the capacity adequacy target differs in each system, and can be more precautionary in one system than in others. Nevertheless, according to national laws, TSOs are responsible for the reliability of the system, whereas the governments (through their relevant ministries for energy) are responsible for ensuring long-term security of supply. In doing so, governments usually rely on long-term system forecasts developed by the TSOs and the national energy regulators, for instance as it is in France.10 This approach is in line with the principle of subsidiarity, which is subjacent to the Article 3(1) of the 2006 Security of Supply Directive: ‘Member States shall ensure a high level of security of electricity supply by taking the necessary measures to facilitate a stable investment climate and by defining the roles and responsibilities of competent authorities ( . . . ).’ And this invites us to consider governments and their delegates, the TSOs, as responsible for the long term

10 In France, the legislative reference is the Law No 2000-108 of 10 February 2000 on the modernization and development of the public electricity service, consolidated version of 1 January 2012 (Loi No 2000-108 du 10 février 2000 relative à la modernization et au développement du service public de l’électricité). Article 1 of the Law states that ‘the public service of electricity aims to ensure the supply of electricity throughout the country.’ A public service contract is then established with the historic company EDF and its independent transmission subsidiary RTE, reflecting the various public service missions. It specifies the concrete specific commitments of RTE. In particular RTE is responsible for the reliability of the electricity system and, as such, it must perform a multi-year forward estimate every two years. Title II, Art 6 of Law No 2000-108 states that the government establishes a multi-year investment program in power generation and that ‘in developing this program, the Ministry for Energy is based ( . . . ) on the multi-year forward estimate prepared every two years under state control, by the operator of the public transmission network.’ The RTE’s forward estimate is exclusively an issue of security of supply and plays, as such, a warning role. Meanwhile the Act gives the Ministry the authority to issue requests for tender from the multi-year investment program. According to Art 8 of Law No 2000-108, the government ‘can intervene, if necessary, on the development of the means of producing electricity through mechanisms of tendering and purchase obligation.’

98

Capacity Mechanisms and Cross-Border Participation

security of supply of their systems, as the TSOs are in last resort responsible for the supply reliability of the latter. Furthermore, the EU approach does not consider the problem of capacity products trade in situations with a risk of congestion of the interconnections to system X during its critical periods. If the probability of congestion is quite high, this makes the bilateral and commercial treatment of cross-border participation deeply problematic. At times of congestion during critical periods of X, no single generator from system Y which bilaterally contracted its capacity with system X, should be considered as having individually contributed to the supply reliability of system X, as opposed to generators of system X. Market coupling ensures the optimization of energy flows from the system with a less critical scarcity situation to the one with a more critical scarcity situation, guaranteeing that the interconnector is consistently congested, but with no relation to the capacity contracts from the former to the latter. When congestion between systems occurs, explicit cross-border participation with trade in capacity rights doesn’t make sense in physical and economic terms. It is the whole of system Y which contributes to the reliability of system X in a statistical way, not the sum of energy flows coming from the generators of Y which participate in the capacity mechanism of X up to the interconnection capacity, and commit to being reliable at the delivery date. During periods of congestion from Y to X, any marginal contributions to X’s supply reliability can only come from X’s domestic generators. What still does make sense is to calculate the probabilistic contribution of imports to the system’s generation adequacy, termed ‘implicit cross-border participation’ without trade of capacity rights. But implicit cross-border participation is considered by the majority of electricity market experts, transmission system operators, and regulators as conservative and even uneconomic.11 This position of vigorously promoting crossborder trade for any electricity-related product, regardless of cross-border congestion, holds little economic weight. I defend, in this chapter, the opposite view that the combination of implicit cross-border participation in the long term and efficient coupling in the short term should bring social efficiency with or without transmission congestion, while this is not the case for explicit cross-border participation when there is congestion between systems during their critical periods. The following three sections focus on the physical and economic fundamentals of bilateral cross-border trading of capacity rights. In particular, I analyse arguments for and against explicit cross-border participation by comparing it to implicit crossborder participation. Section 6.2 discusses the physical and economic relevance of explicit cross-border participation and points out the difficulties in cross-border trade of capacity rights that are inherent to the nature of this product, especially when there is congestion on interconnections between systems during their scarcity periods. Sections 6.3 and 6.4 examine, in two steps, the social benefits of trade in

11 See for example, ACER, Report on capacity remuneration mechanisms and the internal market for electricity, 30 July 2013 (ACER’s Report). ENTSO-E, Communication paper on capacity remuneration mechanisms (WG-RES, June 2012). Eurelectric, Options for coordinating different capacity mechanisms (December 2013, Brussels).

6.1 Elements of EU debate

99

capacity rights. Section 6.3 analyses the efficiency of explicit cross-border participation from the perspective of an individual system of a Member State, provided that the external generators (ie generators located abroad) are able to supply the same capacity product as the domestic generators in this system. Section 6.4 moves to the EU level, in order to identify additional social welfare gains. When the energy markets are fully (physically) integrated, explicit cross-border participation brings economic gains (a) in temporary situations of overcapacities in some systems and (b) in random situations of a time lag between respective critical periods of two systems. However, first, in order to simplify my arguments, I explain the economic concepts of capacity rights (which are quite similar to the term ‘capacity products’ used by the practitioners) and reliability rights related to the two collective goods of adequacy and system reliability (section 6.1.1). Then I propose three important hypotheses to further simplify my discussion (section 6.1.2).

6.1.1 Capacity rights and reliability rights The theory of public economics compares the social efficiency of the different ways of managing collective goods: either by voluntary provision ‘à la Coase’ with the creation of exchangeable property rights between private agents and the entity which offers the collective good; or else by different types of government provisions—pigouvian taxation, specific taxation to fund the production of collective goods, or subsidization of private production of the collective good,—which imply the identification and definition of property rights.12 Adequacy and reliability are two collective goods which are closely related in sequence, as illustrated by Figure 6.1. Adequacy is a long-term insurance that the supply of the power system will be reliable, with the help of new capacity installations and incentives on each equipment committed in the capacity mechanism to be reliable during its critical periods. Attached to the two collective goods of adequacy and reliability, they are exchangeable property rights. ‘Reliability rights’ are rights that are offered every generation unit which is able to adjust their production when producing, or which is in reserve, ready to produce energy and to offer balancing services and ancillary services sold to the TSO, which is in charge of guaranteeing the system reliability to every producer and consumer. Any kilowatt hour (kWh) injected in the system also includes an implicit reliability right. It should be underlined that, if all the reliability rights in a system are in principle tradable with the other systems, the TSO must keep hold of some domestic reliability rights in order to balance the system in the ‘real time’ period of one hour. ‘Capacity rights’ are a property rights on capacity adequacy. This is a collective good under the responsibility of the government, the regulator, and the TSO. Capacity rights are clearly defined by quantity-based capacity mechanisms (as opposed to price-based

12

Richard Cornes and Todd Sandler, The Theory of Externalities, Public Goods and Club Goods, 2nd edn (Cambridge: Cambridge University Press, 1986). Eirik Furubotn and Svetozar Pejovich (eds), The Economics of Property Rights (Cambridge, MA: Ballinger Publishing, 1974).

100

Capacity Mechanisms and Cross-Border Participation Short term:

Real time:

(multi-year)

Day ahead and et Intraday

TSO’s last resort operational action

'Collective goods'/ Policy objectives

Capacity adequacy

Reliability

Reliability & Quality

Players

Power plant investment by market players

Balancing responsible suppliers

Centralized management by TSO

Vectors

Incentives by capacity mechanism

Day-ahead & intraday power exchanges

Balancing mechanism & Bilateral reserve contracts with TSO

Long term

Figure 6.1 Temporal sequences of delivering adequacy and reliability Source: Author’s own illustration.

mechanisms). They represent a forward promise of reliability during their critical period of the delivery year. Generators selected by auction in the capacity auction or, selected by the procurement of obligated suppliers, in the capacity obligation, would typically be required to offer that capacity at the day-ahead scheduling stage and to maintain availability until some nominated point in time before delivery. If the generation in receipt of capacity payments does not clear the dayahead auction, it would be available at the intraday stage to provide capacity or balancing services. In other words, it commits to be available to sell reliability rights in the form of energy or reserves on day-ahead, intraday, and balancing markets. In this case, these rights correspond to forward commitments to be able to commit on the reserve markets or to deliver energy on the energy markets at the delivery date any time during the critical period. In these capacity mechanisms, the capacity right is a complex product. The physical component of the right includes, on the part of contracting generators: a physical obligation of the contracting generator to provide energy during certain critical periods, the definition of these critical periods, an exante certification, an information requirement on outage, and an ex-post control that contracting generators have been available. The same physical component includes, in the case of the bilateral obligation on retailers: an obligation of capacity rights adjusted on the forward peak-load plus a reserve margin, the definition of this reserve margin, and the ex ante and ex post verification by the TSOs before and after the delivery date. In the case of a centralized forward capacity market, it includes the definition of the demand function by the TSO besides the reserve margin.

6.1 Elements of EU debate

101

6.1.2 Three simplifying hypotheses First, the types of capacity mechanisms considered here are quantity-based mechanisms, either capacity auctions13 or capacity obligations.14 These types of mechanisms create clear capacity rights to be traded across borders. I also consider that the capacity product to be exchanged presents a credible guarantee for the reliability of foreign capacity during critical periods of the importing system. It is not the case for capacity mechanisms which have no ex post control of effective reliability of the plants during critical periods and no credible incentive to be reliable via a significant penalty. It is also not the case of wind-power capacity crediting and demand side response products. Second, I only consider capacity mechanisms based on physical rights, and not those based on financial rights, because of the physical reality of the capacity adequacy in particular when cross-border participation is at stake. The reliability options mechanism is first of all a way to protect consumers against price variability and price spikes. It relies on financial contracts with forward coverage, which presents the advantage to guarantee a stable revenue per MW to producers.15 It cannot cope with the physical problem of forward guaranteeing reliability rights supply from one system to another during the critical periods of the latter if there is congestion on interconnection from the former to the latter. Third, I assume that TSOs and regulators have adopted common criteria for capacity adequacy (for instance, setting the same targets for LOLE16) in order to avoid the issue 13

In a capacity auction, the price is set by a centrally conducted auction in which generators bid for (forward) capacity contracts several (usually four to five) years in advance of delivery. Both existing and new capacity providers may participate. Forward capacity contracts might be differentiated between new units (for which capacity revenues could be guaranteed in some way or another for several years), and existing units (for which revenue is only guaranteed for the year of delivery). For more detail on capacity auctions, see section 1.2.3.2 above. 14 A capacity obligation is close to the capacity auction in the sense that it deals with physical rights and that the capacity auction consists in fact in a delegation of the suppliers’ capacity obligation to the TSO, in exchange for the payment of its cost by the suppliers. An obligation is established three to four years in advance for suppliers to sign contracts with new and existing generators. At the delivery date, the suppliers must submit the required number of capacity certificates equal to the peak load of their customers’ portfolios, plus a surplus corresponding to the reserve margin needed for system reliability. It is defined a number of years in advance by the system operator in order to give time for investors to install new peaking units. A secondary market is implemented for marginal adjustments by the obliged suppliers and the committed producers to ensure reliability. Such a capacity obligation is currently implemented in France (see chapter 14). See section 1.2.3.3 above for a discussion on capacity obligations. 15 Reliability option mechanism relies on financial ‘call options’ allocated by auctions. Considering one system, generators who contract with the TSO receive the option premium in exchange for a guarantee that their generation capacity will be available during peak periods (and in case of new investments, guarantee that the capacity will be built). This guarantee is balanced against a capping of the revenue by the option strike price. This aims to guarantee stable revenue streams with energy revenues capped by the strike price and a fixed premium per MW. Considering one system, reliability options can be viewed as an efficient way to encourage investment in peaking units and to provide an incentive to every unit committed in a reliability option contract to be reliable during peak time. But in a multi-system linked by interconnections which may be congested, it is a different problem. We are in a ‘physical right’ world: external generators committed in a neighbouring capacity mechanism than their home one should be exactly those who offer energy and reliability rights to this system at the delivery date. If not, it is an unjustified rent allocation system. 16 LOLE stands for the loss-of-load expectation and is one of the possible indices to measure the level of supply reliability. See section 5.5.1 for an explanation.

102

Capacity Mechanisms and Cross-Border Participation

of free riding.17 Furthermore, if the explicit cross-border participation model is not adopted, I assume that each TSO takes into account the statistical contribution of their neighbouring systems in their own capacity adequacy calculation.18

6.2 Explicit cross-border participation with congested capacity of interconnections Promoters of cross-border trade of capacity rights defend the principle of free exchanges regardless of the situation of the interconnections to the systems in scarcity periods. For them, capacity adequacy of any system is a ‘collective good’ whose supplies could be enlarged to external contributors, even if congestion on interconnections from their system can be anticipated at the delivery date. They assume that the contributions of an external system to its neighbouring system’s generation adequacy could be individualized by candidates coming from this external system to the capacity mechanism auctioning of the latter. By symmetry between systems, their position is equivalent to considering that the adequacy is the same collective good covering the two systems, even if highly probable congestion of the interconnection would make adequacy a different collective good specific to each one. In the following, I detail first the argument that the relevance of explicit cross-border participation in the capacity mechanism of system X will differ in situations with and without cross-border congestion during system X’s critical periods. Second, I argue that, when congestion during its critical period can be anticipated, market coupling offers exactly what it aims at when explicit cross-border participation is passionately advocated, namely that system Y with the least critical scarcity situation during peak periods ensures a contribution to system X’s reliability up to the interconnection capacity, and that this contribution will be determined by the implicit auction of the market coupling in the most socially efficient way, without the necessity of forward capacity contracts auctioning.

6.2.1 Physical and economic fundaments of capacity rights trade Let us again consider the simple example of two interconnected systems X and Y with explicit cross-border participation between them. Ignoring the physical realities of the power system, let us assume that it is possible to identify generators of system Y which contribute to capacity adequacy of system X. This supposition would fictitiously mean that generators of system Y which forward commit to be reliable during X’s scarcity periods to supply energy and reliability rights to this system, isolate themselves from their own system and use a dedicated interconnection line to transport their energy to 17 The subjacent assumption is that Member States do not rely on the more cautious neighbouring systems in ensuring their own security of supply. Accordingly, I assume that different wholesale markets have similar price caps, and that capacity mechanisms in these markets have a similar role of compensation for the missing money resulting from the respective price caps. 18 I make this assumption, because, at present, a number of TSOs do not take into account imports when estimating the necessary reserve margin for their system and establishing their adequacy target. In particular, this is the case of Belgium (see chapter 13), Italy (see chapter 17), and Spain (see chapter 21).

6.2 Explicit with congested capacity interconnections

103

system X. This fiction of traceability can find a physical example with the existing power trade from the giant James Bay hydro plant’s production in Quebec to Massachusetts by a dedicated DC line of 1,100 km to honour a very long-term export contract. Without cross-border congestion, no problem of relevance really arises. This is because generators of system Y which are remunerated in the capacity auction of system X and clear on it can be considered as effectively exporting their energy and reliability rights in relation to the capacity rights they sold in the auction. In terms of the public economics, system reliability as a collective good covers both systems X and Y. The same holds for generation adequacy which relies on the generators’ forward commitment to be reliable and to deliver energy during critical periods of each system, whatever the capacity mechanism they participate in. In case of cross-border congestion between systems X and Y, any generator of Y participating in X’s capacity mechanism, cannot be regarded as improving X’s security of supply at the delivery date. Even if an effective contribution of system Y to the supply reliability of system X can be observed in purely statistical terms, no individual generator of Y can be considered, in the ‘marginalist’ terms of the public economics approach of a collective good supply, as contributing to system X’s supply reliability. Certainly during periods of congestion, trade in energy and reliability rights still exists and electricity continues to flow from system Y to the more stressed system X, which has a higher clearing price. But any marginal contributions to system X’s supply reliability can only come from system X’s domestic generators. This explains the scission of supply reliability as a collective good of purely integrated markets into two collective goods specific to each system.19 The same holds for capacity adequacy, which again is a collective insurance which relies on generators’ individual forward commitments to be reliable at the delivery date. Concretely, generators of system Y which participate in the capacity auction of system X would benefit from undue capacity remuneration because it is not demonstrable that the electricity injected in their home system has been exported to system X among the exported power flows restricted by the congested interconnection. Whatever it would be, to be the closest to this physical fiction of traceability and make credible the system Y-generators’ commitment to be available and produce energy to be sent to system X through the interconnection which can be congested, one minimal condition would be needed. The capacity rights trade from system Y to system X should always be accompanied by firm transmission rights on the interconnection capacity, so that the TSO of system X knows that it can rely on committed generators of system Y during critical periods, in situations when interconnections from system Y to system X could be congested during the latter. At the allocation step of physical transmission rights, the reservation of transmission rights is necessary.20 19 Peter Cramton, Alex Ockenfels, and Steven Stoft, ‘Capacity market fundamentals’, Economics of Energy and Environmental Policy 2 (2), 42–3. 20 In order to set up an import or an export programme exchange, a market party must first acquire transmission rights—the allocation step—and then declares to TSOs the exchanges program it wishes to implement within the capacity acquired—the ‘nomination’ step. The available commercial capacity is allocated through different time frames: a part of this capacity is allocated as periodic Physical Transmission

104

Capacity Mechanisms and Cross-Border Participation

Indeed, as congestion occurs during system X’s critical period, the interconnection capacity has to be shared between ‘normal’ cross-border transactions, not linked to cross-border capacity contracts, and the transactions of sales of external generators committed by capacity contracts to transmit their energy and their associated reliability rights to system X.

6.2.2 Market coupling against the fiction of traceability Transfer of capacity rights from generators of system Y to system X requires firm reservation of the corresponding interconnection capacity on the scarcity periods of system X at the delivery date. Market coupling doesn’t follow this logic, and, if congestion occurs, the day-ahead implicit auction on the interconnection capacity via the market coupling algorithm allocates transmission rights in a socially optimal way, regardless of the type of cross-border transactions involved. Market coupling which covers day-ahead markets and infraday markets, calls into question the relevance of firm reservations of the interconnector’s capacity. It doesn’t guarantee at all that the price offered on the coupled market by generators of system Y committed in the neighbouring capacity mechanism of system X, will be sufficiently competitive to be selected in real time at the delivery date of the forward capacity contracts. Further, the EU Target Model21 includes a ‘use-it-or-lose-it’ rule according to which interconnection capacity reserved in the forward markets is offered in the implicit auction, unless a flow is nominated. So the credibility of forward commitment to deliver electricity and reliability rights to system X via an interconnector is much lowered in the European market coupling model. In fact, market coupling implicitly denies the fiction of traceability of electricity when systems are separated by congestion. However, Baker and Gottstein22 argue that: Under certain conditions, it could have a certain usefulness. External generators selected in a neighbouring capacity auction would typically be required to offer that capacity at the day-ahead scheduling stage and to maintain availability until some nominated point in time before delivery. If the generation in receipt of capacity payments does not clear the day-ahead auction, it would be available into the intraday stage to provide energy or balancing services, despite having given up its reserved interconnector capacity. Moreover, when the external generation in receipt of capacity payments is displaced at the day-ahead stage, then the replacement generation will provide the equivalent capacity across the interconnector. Effectively, the capacity payments would be an ‘insurance policy’ to ensure that the interconnector capability is backed by adequate external generation capacity.

Rights (yearly and monthly PTRs) and the other part is allocated as daily physical transmission rights. There is no capacity reservation for the intra-daily allocations. The PTRs acquired at yearly and monthly auctions and non-nominated are automatically resold at daily allocations. 21 See section 5.3.1, n 13 and the accompanying text. 22 Philip Baker and Meg Gottstein, ‘Advancing both European market integration and power sector decarbonisation: key issues to consider’, Regulatory Assistance Project’s Policy Brief, 3 May 2011, p 43.

6.2 Explicit with congested capacity interconnections

105

But, do we need this insurance policy, even in the crucial case of congestion on the interconnections from system Y to system X? Indeed, market coupling of the dayahead markets and intraday markets guarantees that interconnection capacity is fully and consistently utilized due to market price differentials, when interconnector is congested. Therefore, insurance by procuring capacity contracts with external generators becomes redundant and remunerating foreign generation for their capacity is unjustified. The flows of energy and reliability rights from system Y to system X during system X’s critical periods can be guaranteed by the market game, as long as price differentials reflect higher scarcity in system X. So, the conclusion is quite clear: when market coupling organizes, in an anonymous way, the commercial exchanges of energy and reliability rights inside each system and between systems, any forward commitment of a generator of system Y in system X’s capacity mechanism, is superfluous to system X’s supply reliability. Is there a way to circumvent this problem? In this situation with market coupling, it is conceivable to forward nominate on interconnection capacity and to give them priority access to the interconnector. Their priority access will then be included in the market coupling algorithm, as if they implicitly owned a share in the interconnector’s capacity. However, this solution has two drawbacks. First, it reduces the potential for competition among generators of system Y which sell electricity to system X. Secondly, it challenges the principles of the market coupling algorithm, by giving priority to transactions of external generators committed in capacity contracts, over other cross-border transactions.23

6.2.3 Limits of the solidarity principle in the market coupling The market coupling algorithm allows for maximum solidarity between countries, even in situations of synchronous stresses with first rationings. Indeed the algorithm realizes a mechanical sharing of their respective ‘order-books’. Even if the interconnection capacity is congested, cross-border exchanges are maintained between systems, the sense of the energy flows depending in principle on the highest bid price which clears the respective markets of systems X and Y. As the marginal bid prices in each system could rocket up at some arbitrary levels, the contribution of one to the reliability of the other could be as well from system X to system Y, as from system Y to system X (which is the traditional sense). In fact in the case of respective rationings, the algorithm

23

This type of optimization of the interconnection capacity allocation via the market coupling algorithm doesn’t take place in other market designs like interconnected mandatory pools, without market coupling between systems. This is the case in the US regional markets with capacity mechanisms, for instance between the MISO and the PJM ISO. Transmission rights are allocated there by auctioning. But the fact is that energy exchanges between systems with different situations of capacity and reserve margin are not developed at the level that they should be during respective critical periods because of imperfections in market designs. See Brattle Group, ‘MISO-PJM capacity market seam: preliminary issue description’ (Report prepared for MISO, December 2011). In this case capacity rights sales combined with firm capacity reservation which, allocated by auctioning up to total interconnection capacity, could be a way to efficiently use the interconnections at the delivery date during respective critical periods, and to make credible physical commitment of external generators in the capacity mechanism of the neighbouring system.

106

Capacity Mechanisms and Cross-Border Participation

realizes a physical equalization of the rationing rates.24 This process may lead to a system being forced to have greater rationing. In any event this raises a problem when the adequacy policy is more cautious in one system than in the other with a higher reserve margin and more incentive capacity mechanism. It is also the case if interconnection capacities are large and therefore without congestion between systems, the reliability being the same collective good of the two systems in the terms of public economics. In fact the respective situations of scarcity which oblige each TSO to decide rationing would mean that the ‘system reliability’ offer should be completely taken over by each TSO which means that despite the absence of congestion, this collective good is temporarily separate for the two systems. In both cases, the rule of complete solidarity which is implicit to the market coupling algorithm should be revised in the future. How should energy be allocated in times of scarcity, considering efforts made through heterogeneous margin reserve targets and different adequacy policies?

6.2.4 Comparison with implicit cross-border participation To compare, implicit cross-border participation makes sense during critical periods in all market designs, with or without market coupling. At times of congestion, the contribution of system Y to the supply reliability of system X is taken into account statistically, but with no possibility of identifying the individual generators from whom the reliability rights come. System Y’s contribution to the supply reliability of system X is taken into account in the calculation of system X’s outage probability and its reserve margin target. ENTSO-E is currently developing common methods to evaluate mutual contributions between systems in the EU.25

6.3 Social efficiency of cross-border participation from the national perspective Let us assume that capacity rights can be traded cross-border even in situations of congestion, in the same way as they can be transferred without congestion. What could be the benefits of explicit cross-border participation for an individual system? In order to allow foreign generators to participate in a capacity mechanism on equal footing with domestic generators, two simple conditions need to be fulfilled. First, foreign generators should be able to offer exactly the same capacity products as domestic generators. This means that the services offered by foreign and domestic generators should be the same in terms of the guarantee of reliability and the energy

24 For example, if at a given time, country A has a rationing of 2000 MWh on a load of 10,000 MWh (20%) and country B has an outage of 150 MWh on a demand of 15,000 MWh (rate of 1%), exports will take place in country B to country A until the rationing rates are equal (by 8.6%). This process leads to a country being forced to have a greater rationing. 25 ENTSO-E’s Report, Scenario outlook and adequacy forecast 2014–2030, June 2014 (ENTSO-E’s Report).

6.3 Social efficiency from national perspective

107

sale during critical periods, penalties, rules of certification, and ex-post checking, etc. Secondly, external generators, which participate in a neighbouring capacity mechanism, should opt out from the capacity mechanism in their home system (if such a mechanism exists there) in order to avoid a double capacity remuneration.26 That also means that, in situations of congestion, capacity rights are transferable thanks to firm reservation on the interconnection. Assuming that these two conditions are fulfilled, a system which opens its capacity mechanism to foreign generators (system X) appears at first glance to benefit in two ways. The first benefit is the additional capacity which external generators forward commit to deliver to its system. The second one is the lower cost of its generation adequacy policy due to greater competition in the central auctioning (or the capacity rights purchases by the obliged retailers in the capacity obligation case). However, these benefits appear difficult to materialize for the first and illusory for the second.

6.3.1 Additional capacity contributing to system X’s domestic capacity The first advantage is the supplement of committed capacities to a capacity mechanism coming from external generators. Indeed, the TSO of the importing system X opens the auctioning to every generator of system Y, as well as to the domestic generators to reach its target of capacity adequacy, under the limit of the interconnection capacity for the total of external generators to be selected. So, the number of generators of system Y which are selected by the capacity mechanism of system X, guarantees offers of energy and reliability rights to system X during its critical period. The advantage for the TSO of system X comes from the fact that it adds some guarantee of its system during critical hours by the diversification of the formal commitments. However, this advantage disappears in situations of congestion on the interconnection from system Y to system X. The reason is the same as already mentioned: the impossibility to dissociate an external generator receiving remuneration from the capacity mechanism of system X and to certify that his injections of energy are the ones that moved to system X during critical periods, or else to be sure that this generator would not be displaced by market coupling. Even if an external generator is supposed to offer the same capacity product as an internal generator does, and this product is to respect the same standard of supply reliability inside system X, congestion on interconnections during system X’s critical periods makes the equivalence of capacity products between external and internal generators impossible. Even if we use the fictitious guarantee of transfer by firm reservation of access, and make an external generator pay for it, this does not at all increase the guarantee of system X’s supply reliability at the margin. So, this first advantage only holds when systems and energy markets are fully integrated without congestion between them. But, at the very least, this advantage is

26

It could be possible to avoid an explicit opting out if the respective critical periods of the generator’s domestic system and the neighbouring one (for which the generator applies for a forward capacity contract) are not synchronous.

108

Capacity Mechanisms and Cross-Border Participation

captured by the implicit cross-border participation when the market coupling algorithm organizes the full integration of energy markets.

6.3.2 Lowering the cost of system X’s adequacy policy A priori, if efficient generators and investors exist in system Y, one can easily imagine that those of system Y’s generators who prefer to apply to the capacity mechanism of system X would decrease its clearing price, if we compare it to a situation with implicit cross-border participation and without trade of capacity rights. Moreover, domestic generators of system X would reduce their bid price under the pressure of these external competitors. One should be inclined to consider that, in this way, the cost of the adequacy policy of system X will be reduced for its consumers. But there is no cost reduction by competition. In fact by another way, the cost for the consumers in system X will be greater if there is explicit cross-border participation to pay the capacity rights selling by external generators while there is nothing to pay with implicit cross-border participation. I develop this point below. In the first case, the fact that other generators can add their bids to those of internal generators does not change the clearing price of the capacity mechanism of system X. It will not be lower than the price in a situation without trade and with implicit crossborder participation in system X. The simple reason is that the supply and demand curves in situations with explicit cross-border participation and those with implicit cross-border participation are not the same, as demonstrated in Figure 6.2. It is assumed there that (a) system X has an adequacy target of 100 GW (for an extreme peak-load of 90 GW with a reserve margin of 11 per cent) and (b) the interconnection capacity is such that I estimate statistically the contribution of system Y to the adequacy of system X to be 20 GW. Figure 6.2 presents the different steps of the merit order supply curve of capacity rights in situations with and without explicit cross-border participation. The left-hand graph represents the situation with explicit cross-border participation with a demand of 100 GW and a supply curve including the import of low-cost capacity rights of 20 GW. The right-hand graph represents the situation with implicit cross-border participation with a demand of 80 GW of capacity rights only, and an offer excluding the import of capacity rights. One can observe that in the market equilibrium, the demand curves when they are price inelastic and so are represented as vertical lines, intersect the respective merit order supply curves on the same capacity price step. In the situation of implicit crossborder participation (the right-hand graph), the vertical demand has been displaced by 20 GW on the right, while the higher part of the merit order curve has also been displaced by 20 GW in comparison to the supply curve in the left-hand graph which represents a situation with explicit cross-border participation. Therefore, the clearing price will be the same for the producers in system X. But, in this situation, the cost of the adequacy policy for the consumers will be higher, because the external generators have to be paid by them while they do not displace internal generators. Conversely consumers don’t have to pay external generators in market design with implicit crossborder participation.

6.3 Social efficiency from national perspective Capacity demand w/o cross border contribution

Total capacity demand with crossborder participation Price (€/MW-y)

109

Price (€/MW-y)

Capacity target (MW) Local base-load units

Net Capacity Target (MW) External generators

Local peaking units

Figure 6.2 Comparison of capacity mechanism equilibrium with explicit and implicit cross-border participation Source: Author’s own illustration.

Discussion might be developed about the realism of the capacity rights market equilibrium when the demand function is price inelastic, as is the case here. Indeed, with a price-elastic demand of capacity rights, the equivalence of the equilibrium price on the capacity mechanism between situations with or without cross-border participation, could be challenged. In fact, given the very discontinuous nature of the supply curve of capacity rights in steps of increasing marginal costs, the inclusion of a priceelastic demand function of capacity rights changes little in the market equilibrium comparatively to a situation with a price-inelastic demand of capacity rights. The clearing capacity offer would most probably be a local one with a price bid higher than all the external capacity price bids. To conclude, the capacity price is identical and there is no advantage to enlarging the market playing field to include external generators, in this respect. As a consequence, given that in the system with implicit cross-border participation, consumers do not have to pay external contributors, the cost of the adequacy policy in their system is lower than the same policy with explicit cross-border participation. In any case transaction costs would be higher with explicit cross-border participation than with simpler capacity mechanisms with implicit cross-border participation. Do not forget that explicit cross-border participation requires close cooperation between the TSOs in order to certify generators external to system X which are eligible for crossborder trade and also to check ex post whether they have been in fact available during the critical periods of X. A capacity auction of system X with cross-border trade with system Y would be based on a set of arrangements between generators of system Y

110

Capacity Mechanisms and Cross-Border Participation

and their TSO, and between the TSOs of both systems. This latter set of arrangements (between the two TSOs) is particularly important to properly manage capacity adequacy and supply reliability. To balance this drawback, it would be fair to mention an informational advantage of the explicit cross-border participation. Indeed practitioners are inclined to anticipate some difference between the TSO’s estimation of the statistical contribution of other systems in the implicit cross-border participation and the total of capacity rights which will come from external contributors in the explicit cross-border participation. Indeed, any TSO tends to be conservative in his approach to external contributions. So at the end of the day, with an explicit cross-border participation, the total capacity of selected external generators could be higher than the statistical contribution estimated by the TSO if the cross-border participation is implicit.

6.4 Social efficiency of cross-border participation from the EU perspective A national perspective only considers the interest of one system. As just considered in the former section, it is only necessary to suppose that an external generator to a system X is able to offer a capacity product which is similar to the product requested from the internal generators of the system X. If we jump to the EU-wide level, we observe that the individual system approach ignores two issues of EU-wide social efficiency. First, the eventual social cost of no trading of capacity rights and the social benefit of exchanges of capacity rights for the whole set of systems. Second, in case of adoption of different capacity mechanisms, the distortive effects that the external trading of different types of capacity rights causes. Concerning these two different issues, the problem is again different if the systems are fully integrated, or if they are separate by congestion during their critical periods because exchanges make economic sense in the first case. Therefore, one should analyse the issue of social efficiency in the two situations of interconnection for the more stressed system— not congested and congested.

6.4.1 The defence of implicit cross-border participation The Commission assumes negative effects of implicit cross-border participation and no capacity trading in its Generation Adequacy SWD. As already quoted in section 6.1, it has two concerns which can be challenged. First, it notes that ‘a mechanism which excludes cross-border participants could result in new generation capacity displacing imports.’ Second, it states that the statistical contribution approach ‘( . . . ) would undermine the financial viability of generation in other Member States and could have a negative impact on regional security of supply.’27

27

Generation Adequacy SWD (n 6) p 28.

6.4 Social efficiency from EU perspective

111

The first concern does not take into account the difference between a capacity mechanism without import of capacity rights (implicit cross-border participation) and a capacity mechanism with imports (explicit cross-border participation). We have seen in section 6.3, that if there is explicit cross-border participation, the demand of capacity rights in the capacity mechanism covers the whole adequacy target. If there is only implicit cross-border participation, the demand of capacity rights only consists of the net demand, after subtraction of the external statistical contribution. So, with implicit cross-border participation, there is no displacement of imports by new internal generation capacity because the adequacy target is calculated by taking into account the external contribution of energy imports during scarcity periods. The difference is that no external generator gets revenue from the capacity mechanism of a neighbouring system. To challenge the second concern, I put forward three arguments. First, eventual investors can collect capacity revenues from their domestic capacity mechanism, if such a mechanism exists. Second, capacity revenues from other capacity mechanisms with explicit cross-border participation will be uncertain (as are those that can be expected from their domestic capacity mechanism), and certainly not sufficient to trigger investment decisions from external generators. In other words, given that a capacity mechanism has, as a first aim, to encourage investment in new capacities, it is not these occasional revenue differentials between the internal and the foreign capacity mechanisms which will trigger an investment decision in the neighbouring systems. Consequently the regional security of supply would not be altered by the preference of some systems to take into account the statistical contribution of the other systems (rather than allowing bilateral cross-border transactions) because investment incentives in the external systems are not affected by the statistical contribution. Investment will be made in each system under the incentives of the revenues caused by price spikes of the energy market and those of the forward annual capacity price of the domestic capacity mechanism, when the capacity becomes tight. Third, the implicit cross-border participation allows a more programmatic approach with a much greater chance to reach the standard of adequacy at the level of a national system than the EU-wide adequacy approach. This latter is wrongly supposed to be more efficient because it is more market-based. In the case of congestion on the interconnections from system Y to system X, a new argument adds to these three considerations on implicit cross-border participation versus explicit cross-border participation. Any new capacities invested in system Y with the incentives of complementary revenues from the capacity mechanism of system X do not contribute at all to the security of supply of system X. The problem of the capacity adequacy of system X has to be solved with incentives which are internal to system X when there is congestion on interconnections from the neighbouring system Y to system X during critical periods of the latter.

6.4.2 Which social benefits of an explicit cross-border participation? We have again to distinguish between situations without and with congestion on interconnections between systems.

112

Capacity Mechanisms and Cross-Border Participation

6.4.2.1 Limited social benefits of explicit cross-border participation when there is no congestion Assuming the most favourable situation where there is no congestion in times of scarcity between the two systems and where countries have the same criteria of adequacy, what could be the economic gains from the trade of capacity rights compared to the case of implicit cross-border participation? Eventual social benefits of explicit cross-border participation by comparison to implicit cross-border participation could be envisaged: the postponement of closures in the systems in overcapacity and the competitive advantages of a system in developing peaking units (gas turbines, etc), hydro plants (pumping storage, reservoirs ‘uprating’) or demand-response programs. I distinguish between (a) short-term and (b) long-term benefits. With respect to (a), short-term economic gains from the trade of capacity rights will emerge in two situations. First, we can find advantages in situations of transitory overcapacity in system Y by comparison to the case of implicit cross-border participation. Here, an exchange of forward capacity rights from system Y in overcapacity to system X with a capacity mechanism, which is in scarcity, allows us to postpone the construction of new reserve units in system X. At the same time, it may provide some revenue for the equipment about to be closed in system Y in overcapacity, when their annual energy revenues are too small to cover their fixed operating costs, as we have observed, since 2012, in some EU countries. They would not have such revenues with implicit cross-border participation and no trade of capacity rights.28 But, we should not forget that these situations are temporary, and could be shortened by private decisions to close some plants, by variable renewables capacity increases in system Y which create a new need for reserves, or by unanticipated higher economic growth which will increase peak-load to an unanticipated level. Second, there are advantages to trade in capacity rights if there is a ‘structural’ decorrelation between two systems’ randoms concerning their respective total loads and their intermittent renewables production. At the multi-system level, such structural decorrelation leads to differences in the respective critical periods. This makes overproduction regularly appear in, let us say, system Y, which allows it to export energy with a significant degree of probability to system X during the critical periods of the latter, up to the interconnection capacity, and vice versa. So there is a mutual exchange of services, which could help to lower the need for reserves in each system. But the same benefit exists with implicit cross-border participation. So, what will be the difference with explicit cross-border participation? None in terms of social benefits. The only difference will be that variable generators of system Y will be remunerated up to their capacity credit by the capacity mechanism of system X for their stochastic contribution to the security of supply of system X. But, there is a de facto limitation

28 Notice that the capacity revenues for the older and less efficient equipment of system Y could come from the capacity mechanism of system X as well the capacity mechanism of system Y. Indeed, if old units of system Y do not manage to bid more efficiently on the capacity mechanism of system X than more recent generation units of system Y, they will replace these latter ones on their domestic capacity mechanism in system Y, given that those units of system Y, which have been selected on the capacity mechanism of system X should have opted out from the capacity mechanism of system Y.

6.4 Social efficiency from EU perspective

113

because renewable electricity production is variable. So the advantage of trading forward capacity rights between systems related to their respective scarcity periods is fleeting because of this variability. Consequently the TSO of system X will probably be very conservative in the attribution of capacity credit to the variable RES-E generators of system Y, because system X could confront the risk of very low RES-E production in system Y when system X is in its critical periods. With respect to (b), economic gains will also emerge from the perspective of longterm market equilibrium, the only one that, in fact, really matters for a capacity mechanism, when new capacity development is at stake in some systems. There could be long-term gains from trade in situations where system Y could be much more socially efficient (by reference to the optimal mix of the two integrated systems) in building new peaking units among which new pumping storage or uprating of hydro reservoirs, or in developing larger demand response programs than in the neighbouring system X. So, in system Y, whatever the anticipated situation of system adequacy and eventual introduction of a domestic capacity mechanism, generators can decide to build such units to sell capacity rights to the neighbouring capacity mechanism in system X with the possibility to compare anticipated capacity revenues from capacity mechanism of system Y and the neighbouring capacity mechanism of system X. But, this hypothesis of competitive advantages on peaking unit construction costs is not realistic for combustion turbines because they are standardized technologies with the same cost from one system to another.29 And investors would do better to build peaking units in the neighbouring system to avoid transmission cost. Another case of a long-standing advantage is the potential to increase incentives to invest in peaking hydro plants in neighbouring countries. It concerns neighbouring countries endowed with sites allowing the development of new pumping storage or uprating of hydro reservoirs. These technological opportunities are both dedicated to answering flexibility and forward capacity needs. But the present experience of Alpine pumping storage projects shows that, for the first part of expected revenues from flexibility services which are based on arbitrage in energy markets between different hours in the day or the week, prospects of revenue are fleeting. Indeed, under the effect of large-scale photovoltaic (PV) generation in the region, reduction of spreads between day and night prices has deterred investment in several projects. A final remark to further reduce the scope of this supposed advantage, is that a capacity mechanism of any system is unable to give long-term visibility to capacity revenues in order to limit the risks of investing in new capacities, whether it is inside the system or for external generators. To conclude, if explicit cross-border participation is compared to implicit crossborder participation, there is no stable short-term or long-term competitive advantage in the trading of capacity rights, only occasional advantages due to overcapacities here 29 We could consider a particular case with hydro-dominant countries with seasonal or weekly storage capacities (Austria and Switzerland for Germany and now Norway for the Netherlands, Germany and possibly for the UK), but these assets already exist and new developments are not at stake. In this case, exchange of capacity rights from these systems to other systems would depend on the existence of water inflows over the statistical average in the dam during peak periods. But this does not allow firmness of forward commitments to the neighbouring systems.

114

Capacity Mechanisms and Cross-Border Participation

and there. As the main objective of a capacity mechanism is to correct the market failure in the case of investment in new capacities, it is necessary to keep in mind that short-term benefits of trading capacity rights should be balanced with the longer-term cost of having underinvested in the systems without overcapacity. Indeed, there is some possibility that congestion may appear on an interconnection, because of increased energy flows coming from system Y in overcapacity to system X during its critical period. At the end of the day, the potential advantages of allocating capacity revenues to external generators in situations of no congestion during respective critical periods are not conclusive. It will then be underlined again that improvement of market integration has much higher positive effects than capacity rights trading.

6.4.2.2 No social benefits of explicit cross-border participation, in case of congestion The discussion in the previous sections holds only if there is no congestion on interconnections between systems during their respective critical periods at the delivery date. Consider the case of two systems. If congestion exists or might appear between them during their critical periods, adequacy is no longer a collective good which is common to the two systems, but should be seen as two separate collective goods. So, trade of capacity rights is problematic if there is a risk of congestion. On the one hand, it could help to postpone closure of plants in system Y in overcapacity from revenues coming from the capacity mechanism of system X, which is in stress. But on the other hand, this does not help to increase capacity in system X. Indeed, generators external to system X have captured part of the capacity revenues of the capacity mechanism of system X, as their bids also lower the capacity price on this capacity mechanism. Consequently, internal generators have not invested enough. With implicit crossborder participation in this case of congestion, internal generators have clearer incentives to invest because the capacity price is higher, given that they are not exposed to the competition of existing generators of system Y. Another problem already mentioned is that nothing guarantees that energy exports from system Y to system X during critical periods correspond to the units committed in the capacity mechanism of system X, during system X’s critical period when there is congestion. It is an undue surplus for them which is paid by the consumers of system X. It is both an issue of efficiency and an issue of equity for the two systems.

6.4.3 Distortive effects of cross-border trade with different capacity mechanisms There are many reasons to anticipate a proliferation of disparate capacity mechanisms with differences in their attractiveness for investors in the different electricity systems.30 For that, all other things being equal (in the same situation of capacity scarcity 30 In this section, I refer to different capacity mechanisms (see section 1.2.3 for an overview), while up to now, I have only referred to capacity auctions and capacity obligations (the auctioning of forward capacity

6.4 Social efficiency from EU perspective

115

of each system), the most attractive capacity mechanism would offer more transparency and more revenue stability from one year to the next. This will de facto install competition between systems to attract investors in peaking units as the fiscal competition plays between Member States to attract foreign firms.

6.4.3.1 Comparison of implicit and explicit cross-border participation without congestion Let us consider the case of implicit cross-border participation between two systems, X with the most profitable capacity mechanism and Y with the least profitable one. The differences in adequacy approaches alter the long-term optimum of each individual system, considered as if they were separate. Indeed, system X attracts the peaking unit investment, and consequently will sell more and more energy to system Y during its critical period. It is not problematic until congestion emerges on the interconnection from system X to system Y during system Y’s critical periods. After establishment of the congestion, the markets will no longer be fully integrated during these critical periods. System Y will have to suffer higher energy and capacity prices and accept some outages during its critical period after congestion occurs. Prospects of higher capacity revenues and scarcity rents on the energy market will attract investors to install peaking units in system Y.31 For the time being, congestion on the interconnection from system X to system Y will disappear, and energy markets will be again integrated for a period depending on the demand growth rate in the X and Y systems.

6.4.3.2 Comparison of implicit and explicit cross-border participation with congestion between systems Let us put aside the lack of economic and physical rationale of capacity right exchanges between two systems when congestion arises. Allowing cross-border trade of capacity products adds a new dimension in the competition between electricity systems to attract investment in new capacity for the objective of adequacy: They need to get sufficient capacity rights to reach their adequacy target at the lowest cost, not only from attracting more investors to install new capacities at home, but also by getting part of them from lower external bidders to their respective capacity mechanisms. Explicit cross-border participation amplifies the economic distortions that result from the adoption of different capacity mechanisms in neighbouring systems versus implicit cross-border participation. As far as capacity adequacy is concerned, and because there is no congestion from one system to another during the critical period of the latter, there is, accordingly, no problem in adequacy of the integrated systems.

contracts, or else the bilateral obligation), which both clarify the property rights on capacity in the most relevant way. 31 Mauricio Cepeda and Dominique Finon, ‘Generation capacity adequacy in interdependent electricity markets’ (2011) Energy Policy 39 (6), 3128–43.

116

Capacity Mechanisms and Cross-Border Participation

If there are anticipated congestions on interconnections between systems, once again investment in peaking units goes to systems where the capacity mechanism is the most profitable (system X), all other things being equal. Let us suppose that system Y is stressed in terms of capacity. Fortunately, congestion will play to encourage investment in the most stressed system by the misalignment of prices in the two energy markets during the critical periods of the stressed system Y and by the increase of the capacity price of its capacity mechanism. So the scarcity rents will be more important, while the capacity revenue from this capacity mechanism is attractive. Both will trigger decisions of investment in new capacities until the subcapacity of system Y is corrected. But on this point, there is a difference between the impacts of implicit cross-border participation and explicit cross-border participation. It comes from the fact that explicit cross-border participation creates a discrepancy in the attractiveness of capacity mechanisms not only for investors, but also for any generators external to the system with the most attractive capacity mechanism, all other things being equal. When system Y, with the less attractive capacity mechanism, is closer to scarcity than system X, this means that, at a given moment, because the capacity price will increase in the capacity mechanism of system Y, the generators of system X will begin to be attracted by the capacity mechanism of system Y rather than by their usually more favourable capacity mechanism. In addition, the generators of system Y, usually also attracted by the capacity mechanism of system X, will prefer to bid in their domestic capacity mechanism. Until now nothing is anomalous. But, the problem comes from the fact that this switch occurs at a capacity price level of system Y which is lower than the one that would be seen if the capacity mechanisms were identical, and with a delay. Generators of system Y stay too long on capacity sales to the capacity mechanism of system X because the price of the capacity mechanism of system Y starts to increase from a lower level than it would be if the capacity mechanism of system Y was identical to the capacity mechanism of system X. This will also deter investors in system Y from ordering new equipment during a longer time that it would be with implicit cross-border participation.

6.4.4 Efficient combination of market coupling and implicit cross-border participation In fact, efficient market coupling combined with implicit cross-border participation should bring the advantages that proponents of explicit cross-border participation seek, but they forget to take into account all the benefits of market coupling. Better integration of day-ahead, intraday, and balancing markets mutually reinforces the reliability of each system, and beyond that, the long-term supply reliability insurance. In case of congestion, market coupling guarantees the maximization of the benefits of energy and reliability rights trade between the systems provided that the interconnector capacity is fully and consistently utilized, due to market price differentials. Pooling the flexibility resources via the extension of balancing zones and the better integration of intraday markets, should indeed moderate the expense of new back-up of

6.5 Conclusion

117

large scale wind and solar production. The more important the area of reliability rights, the less the balancing need of each system will require internal adjustment and operating reserve services. Moreover, as an effect on capacity adequacy, the less reserve margin for the security of long-term supply is needed, provided that interconnection capacities are there. So, the issue of integration of energy and balancing markets is of prime importance for the increase of supply reliability in each system and, when addressed, this issue reduces the stake of capacity rights trading while it is so common to rationalize it in the name of increasing supply reliability in the different systems.

6.5 Conclusion This chapter has examined the reality of the advantages of explicit cross-border participation and the drawbacks of implicit cross-border participation, by distinguishing situations without congestion between systems from a situation with congestion during critical periods. The EU approach promoting explicit cross-border participation considers that excluding explicit cross-border capacity from participating in a capacity mechanism is considered to have serious distortive effects on long-term competition, as it doesn’t explore the advantages of multi-system competition. This chapter examines the advantages of explicit cross-border participation by distinguishing between situations with and without cross-border congestion during critical periods. This distinction is relevant, as congestion separates between countries markets for reliability rights and adequacy products during critical periods, and suppresses any economic and physical relevance of a capacity commitment from cross-border participants to a national capacity mechanism. From a national market’s perspective only, there could be two potential advantages of explicit cross-border participation in a capacity mechanism. First, cross-border capacity adds to the committed capacities from national resources. Second, crossborder participation means greater competition, lowering the cost of the system’s generation adequacy policy. The former point makes, however, no sense when there is congestion on the interconnection. The latter point is, however, illusory because the clearing price of capacity in the capacity mechanism would be the same with or without explicit cross-border participation. Moreover, whenever congestion occurs, the benefits of explicit cross-border participation will not materialize. From a broader, multi-system perspective, explicit cross-border participation can lead to social efficiency gains, especially when electricity markets with long-standing overcapacity are linked to those with tight situations during critical periods, or with markets predominantly reliant on hydro resources. However, yet again, congestion counteracts the benefit of cross-border participation. In any case, irregular revenue does not encourage new investment in either market. Furthermore, the exchange of capacity rights between markets with different capacity mechanisms constitutes a greater market distortion than implicit cross-border participation with no trade of capacity rights. Congestion complicates this scenario even further, as it delays the price signal of capacity scarcity in the system with the least attractive capacity mechanism, in terms of revenue and risk management.

118

Capacity Mechanisms and Cross-Border Participation

To sum up, external contributions of other systems to the adequacy of one system cannot be managed by bilateral forward capacity transactions between systems because, beyond the absence of clear advantages, the probability of congestion on interconnections in respective critical periods suppresses any economic relevance to the exchanges of capacity rights. Now that we are faced with the increasing physical and technological complexity of the systems through the development of large-scale variable RES-E capacities, de-correlated thermo-sensible loads, and new electricity usages with an uncertain load profile such as electrical vehicles, the anticipation of capacity margin needs and possible contributions of external generators will be increasingly difficult. The pragmatic approach of implicit cross-border participation should be chosen to manage the interaction of the respective capacity adequacies of neighbouring systems. A reasonable approach is to simply use statistical contribution during critical periods, commonly estimated by the ENTSO-E modelling approach. It is the least inefficient method to decompose the extremely entangled adequacy issues of systems in physical interactions, into a series of manageable subsets. By this pragmatic method, we can avoid expected and unexpected inefficiencies, as well as costly informational infrastructures and huge transaction costs to manage the exchange of capacity rights.

7 The System Adequacy Problem Lessons Learned from the American Continent Carlos Batlle, Paolo Mastropietro, Pablo Rodilla, and Ignacio J. Pérez-Arriaga1

7.1 Introduction The main reason behind liberalizing the power generation activity was to promote economic efficiency at all levels, in the short term (at an operational level), but especially in the long term (at capacity expansion level). This belief was based on fundamental economic theory, which asserts that the short-term market marginal price is all that is needed to remunerate the generators to lead the system expansion towards an optimally adapted generation mix. This market mechanism to rule short-term operation was also seen as the required level playing field to attract new investors. However, from the outset, ever since Chile restructured its power sector with its pioneering reform in 1982, the ability of short-term marginal prices to provide sufficient incentives for investment in generation was called into question in a number of countries in which liberalization was implemented. This was especially the case in the American continent: most Latin American countries (with the exception of Brazil) and some power systems in the United States (eg PJM),2 introduced in their original market designs some sort of capacity mechanism (a capacity market or payment, or sometimes both), aimed to complement short-term marginal prices with remuneration for the available capacity. On the other hand, the majority of European countries (once again with some exceptions, namely Spain and Ireland) followed the so-called energy-only market approach, refusing to implement any explicit capacity mechanism. However, the majority of them employed implicit and subtle regulatory safeguards regarding security of supply.3 In the years following the implementation of the first market-based mechanisms, it has become evident that the theoretical premises, through which the market alone would

1

The authors would like to thank the Fondazione Centro Studi Enel for supporting part of this research. PJM stands for Pennsylvania-New Jersey-Maryland Interconnection LLC (Mid-Atlantic region power pool). Since its establishment in 1927, PJM has continued to integrate additional utility transmission systems into its operations and currently covers all or parts of Delaware, Illinois, Indiana, Kentucky, Maryland, Michigan, New Jersey, North Carolina, Ohio, Pennsylvania, Tennessee, Virginia, West Virginia, and the District of Columbia. 3 Carlos Batlle and Pablo Rodilla, ‘A critical assessment of the different approaches aimed to secure electricity generation supply’ (2010) Energy Policy 38 (11), 7169–79. Examples of regulatory safeguards are represented, for instance, by the presence in the market of large state-owned (or controlled) companies, or by the long-term procurement of reserves to be used during scarcity conditions, either from the generation side or from the demand side (through auctions for interruptible demand). See Part IV (chapters 12–22) for country examples. 2

120

The System Adequacy Problem

provide the optimal investment signal, are unfortunately absent in practice in the vast majority of cases.4 As a result, capacity mechanisms are currently under discussion in almost all of the countries that initially opted for an energy-only market design. Today, this issue is at the core of the regulatory debate in Europe.5 Italy and Portugal have recently redesigned their regulatory schemes in this direction6 and France7 and the UK8 have officially announced the implementation of mechanisms of this nature and are currently in the process of developing the design details. In Germany, the government has publicly announced that it is evaluating the different alternatives.9 On the other hand, over the past decade, the countries that originally introduced a capacity mechanism have improved their schemes. In North America, PJM fixed most of the flaws in its original capacity market design during the implementation of the new Reliability Pricing Model in 2008 (eg there was neither a lag period nor a long-term contract duration),10 and a new capacity mechanism based on auctions (the so-called Forward Capacity Market) was implemented in New England.11 However, it is in South America where most of the activity on this issue has taken place. Here, the fast-paced growth in electricity demand and the difficulties of mobilizing financial resources have stressed the importance of properly designing a mechanism that attracts an adequate level of investment. At the beginning of the twentieth century, several South American countries suffered serious power shortages, which were attributed to deficiencies in the original market designs, and in particular the initial capacity mechanisms.12 These circumstances resulted in the implementation of a second wave of reforms, which significantly modified the old schemes. All these new mechanisms intend to reduce the investors’ price and regulatory risks, through the auctioning of long-term contracts, in order to hedge at least part of their remuneration. Nevertheless, despite this apparently common characteristic, these long-term auctions present significant differences in terms of several relevant design elements to be defined 4 Pablo Rodilla and Carlos Batlle, ‘Security of Generation Supply in Electricity Markets’ in Ignacio J. Pérez-Arriaga (ed), Regulation of the Power Sector (London: Springer-Verlag, 2013) pp 581–622. See Part IV (chapters 12–22) for country examples. 5 See discussion in chapter 1. For the Commission’s approach to capacity mechanisms, see, in particular, European Commission, Consultation Paper on generation adequacy, capacity mechanisms and the internal market in electricity (the 2012 Consultation Paper), available at http://ec.europa.eu/energy/en/ consultations/consultation-generation-adequacy-capacity-mechanisms-and-internal-market-electricity, accessed 1 February 2015, and Commission staff working document, Generation Adequacy in the internal electricity market—guidance on public interventions, 5 November 2013 (Generation Adequacy SWD). 6 Autorità per l’Energia Elettrica e il Gas (AEEG), ARG/elt 98/11, Criteri e condizioni per la disciplina del sistema di remunerazione della disponibilità di capacità produttiva di energia elettrica, ai sensi dell’articolo 2 del decreto legislativo 19 dicembre 2003, n. 379. See chapter 17. 7 RTE, French capacity market. Report accompanying the draft rules (Mécanisme de capacité: proposition de règles et dispositions complémentaires), 9 April 2014. See chapter 14. 8 DECC, Implementing electricity market reform (EMR)—finalised policy positions for implementation of EMR, June 2014 (Implementing EMR). See chapter 22. 9 Peter Cramton and Axel Ockenfels, ‘Economics and design of capacity markets for the power sector’ (2012) Zeitschrift Für Energiewirtschaft 36, 113–34. See chapter 15. 10 John D. Chandley, ‘ICAP reform proposals in New England and PJM’, LECG, Report to the California ISO, September 2005. 11 Peter Cramton and Steven Stoft, ‘Forward reliability markets: less risk, less market power, more efficiency’ (2008) Utilities Policy 16 (3), 194–201. 12 Carlos Batlle, Luiz A. Barroso, and Ignacio J. Pérez-Arriaga, ‘The changing role of the State in the expansion of electricity supply in Latin America’ (2010) Energy Policy 38 (11), 7152–60.

7.3 Target market

121

by the regulator. The proper identification and analysis of these design elements are two major objectives of this chapter. After two decades of our involvement in the development of long-term auctions across the American continent, we would like to share some of the key regulatory lessons that have been learned in the process. This chapter does not describe the design of the long-term capacity auctions in North and South America, country by country,13 but rather, it identifies the main design elements of these mechanisms and defines guidelines for their determination. This analysis should provide a valuable tool to those regulators, in Europe and elsewhere, who are presently designing or planning to introduce a capacity mechanism in their power sectors. The chapter is organized as follows. Section 7.2 presents the main design elements of the long-term auctions implemented in the American continent. Sections 7.3 to 7.7 analyse these elements one by one, studying the impact of their determination on the performance of these mechanisms and, based on this, providing guidelines for all the countries where similar schemes are currently under discussion. Section 7.8 concludes the chapter by collating high level recommendations.

7.2 Design elements Capacity mechanisms based on auctions can be properly analysed by dismantling and assessing their various design elements (eg the target market, the level of centralization, the lag period, or the contract duration). However, those design features that are often undervalued and considered as minor or secondary (as for example, performance requirements, indexation, warranties, contract type, or auction format) can also significantly affect the final result of the mechanism. These elements are not always independent and, in any case, their design should not be decided in isolation. Rather, they should be seen as pieces of a puzzle, to be modelled in a way that allows them to fit together, in order to form a robust regulatory instrument that is as effective and efficient as possible. In the following sections 7.3 to 7.7, the design elements will be studied one by one, characterizing the potential impact that each of the available choices might have on the auction performance and providing guidelines for their determination. Nonetheless, since auctions must be tailored first to the peculiarities of each power system and ultimately to the regulatory and policy objectives pursued, it will not be possible to provide general guidelines which may be valid for every condition.

7.3 Target market 7.3.1 The buying side When designing a long-term electricity supply auction, or more generally a capacity mechanism based on auctions, the first element which must be specified is the type of 13 A review of this kind is already available in literature, see Luiz T. A. Maurer and Luiz A. Barroso, Electricity Auctions: An Overview of Efficient Practices (World Bank Publications, 2011).

122

The System Adequacy Problem

demand that will be involved in the transaction, ie defining who the buyer is, or on whose behalf the regulator is buying. The main options are the following. First, buyers can be regulated customers supplied by distributors or more generally regulated retailers (captive demand). Secondly, buyers can be free customers, usually large users, who are eligible to procure electricity independently (free demand). Lastly, the entire system demand can buy. An additional alternative in this last case is to allow all of the demand to participate in the auction, permitting consumers to opt out, depending on the clearing price, by offering demand reduction products. When implementing this option, it is important to allow demand response bids only from those agents for whom it is possible to guarantee that the demand reduction is effective.14 Penalties for underperformance, similar to those applied to generating units (described below in section 7.6.3) should also be applied. This approach has not been considered in South American long-term auctions, but it has been implemented in the Forward Capacity Market of the ISO-NE, where demand response can bid in the auctions,15 and in the PJM Reliability Pricing Model, where demand response is also accepted as capacity services provider.16 In these cases, capacity auctions have proven to be capable of involving demand response resources, even if this did not result in a full exploitation of demand response potential. This design element is significantly affected by the regulatory objectives of the auction. For example, long-term electricity auctions, originally introduced in order to solve problems related to system adequacy and system expansion, are utilized in the South American context to achieve a secondary objective: hedging the end-user default tariff price. This is the reason why in Brazil, Chile, and Peru the auctions cover only the captive demand, although in the Brazilian case free demand is also required to cover 100 per cent of its requirements through long-term contracts. In these countries, distribution companies (the regulated retailers) are mandated to take part in the long-term auctions, so that they can also set stable default tariffs for their customers for a significant period of time via these mechanisms. Considering now only the original objective of capacity mechanism based on auctions (that is, solving system adequacy problems), the decision about who has to buy leads us to the following question: who benefits from system adequacy? Today, long-term security of supply is still a public good that benefits all the consumers of the power sector, so for now it seems the only correct answer to this question would be the entire demand of the system. Any other arrangement would create an evident situation of free-riding, because some users will be taking advantage of the system adequacy without paying the associated costs. In this context it is important to bear in mind that the South American long-term auctions we discuss here mainly aim at providing investors with a hedge to cover their regulatory and

14 Hung-Po Chao, ‘Demand response in wholesale electricity markets: the choice of customer baseline’ (2011) Journal of Regulatory Economics 39 (1), 68–88. 15 Independent System Operator of New England, Forward capacity market rules (FCM 101), as presented in 2014. 16 PJM Capacity Market Operations, Manual 18 on PJM capacity market, Revision No 25, released on 30 October 2014.

7.3 Target market

123

market risk, so the part of the system demand which actually buys in these auctions bears the related risk premium.17 Since the charge necessary to guarantee the capacity expansion is one of the costs that comprise the total cost of the system, if it is paid only by a part of the demand, this can be seen as a cross-subsidy from the customers who must take part in the long-term auction, to those who do not have to participate.18 From this point of view, the choice made in the Colombian and North American auctions, in which the regulator procures some sort of ‘reliability product’ on behalf of the whole spectrum of consumers, seems to be the most adequate. The most recent Argentinian regulation addresses the same problem in a clearly different way. Following the economic crisis of the early 2000s and the ensuing energy scarcity, the government decided that regulated consumers had full priority of electricity supply and that non-regulated users had to cover their own expected capacity requirements through auctions. To some extent, it can be stated that there is also a freeriding issue with this approach, but in this case in the opposite direction, since free demand bears the cost of attracting new investments, and captive consumers benefit from it.

7.3.1.1 Level of centralization The level of centralization of a long-term electricity auction can refer both to the overall auction process and to the demand forecasting. In the former case, the regulator can either organize one centralized auction or assign this task to other agents involved in the auction (free demand or distributors on behalf of their regulated demand). In the latter case, the regulator either calculates the amount of capacity to be procured or leaves this task to the previously mentioned agents. In the American context, different approaches have been applied. Colombia opted for a completely centralized approach, both in terms of the overall auction process as well as demand forecasting, while the Chilean and Peruvian schemes are fully decentralized.19 Brazil has centrally-managed auctions, but the quantity that needs to be 17 In the South American context, an argument commonly used to justify the decision to lay the obligation to buy in the auctions only on regulated consumers, is that a contract with a distribution company (in their role of regulated retailer) as counterparty reduces the investors perception of credit and regulatory risk, which would be higher if the counterparty is just ‘the whole system’ (in the form of a regulatory commitment, the capacity payment, instead of a contract) or the free consumers. 18 This approach would give a justification for the regulator to shed the load of free customers in case of a future scarcity event, therefore eliminating the concern of free-riding. However, the problem is that it is not always technically feasible to discriminate between categories of consumers when cutting the supply during an emergency event, so a certain level of free-riding will always be present. It is worth noting that the whole discussion may change in the future. New concepts such as smart metering principally allow that security of supply can be turned into a private product (where those not paying for security of supply may be offered a lower level of service). Whether regulators and politicians would allow those consumers who did not pay to be switched off in instances of generation capacity shortfall would be a key question in such context. If politicians would not allow this then we are back to the free riding problem (even if technically we could discriminate consumers in their security levels). 19 Actually, Peru has implemented two different auction schemes. The first to be introduced (auctions under the framework of Law 28.832) only targets distribution companies supplying captive demand and follows a fully decentralized approach. However, a secondary mechanism (managed by the Proinversión

124

The System Adequacy Problem

procured is decided by distribution companies. The ISO-NE and PJM capacity mechanisms can be considered as completely centralized auctions. There are several advantages to a centralized auction approach.20 The most important feature of a centralized auction is that it allows for the full benefit of exploiting economies of scale in electricity generation. If all the agents involved in the mechanism organized independent auctions, the amount of capacity/energy to be procured by each one of them would be too small to justify the construction of a new plant, especially if large hydropower projects are considered. In South America, other advantages of centralized, public, and therefore transparent auctions derive from the specific structure of South American power systems. Only a few countries in the region have unbundled distribution from retailing and the most common market characteristic is to have local distribution monopolies, which are in charge of purchasing electricity for the regulated customers connected to their distribution networks, ie acting as regulated retailers. Furthermore, where this configuration is in place, the central concern is that the unbundling of generators and these distributors/regulated retailers is insufficient at the very least, and often non-existent. Within this structure, vertically integrated distribution companies, acting as regulated retailers, not only have no incentive to design a mechanism that results in the minimization of prices for the end consumers, but they could even arrange the auction process in order to procure their entire future supply needs from the generation part of the same company, at a non-transparent price. Within this structure, a centralized auction minimizes the risk of competition infringements by vertically integrated companies.21 Moreover, this design also ensures that all the regulated demand faces the same energy price, and fulfils the equity principle of tariff design. In fact, in those systems where the auctions are decentralized, smaller distribution companies are commonly exposed to higher prices. There are concerns regarding vertical integration and entry barriers, not only in South America but also, to some extent, in North America and Europe. As it has already been observed in the UK,22 vertical integration between generation and retailing reduces the liquidity in the market and acts as an entry barrier to new competitors. In fact, incumbents plan their capacity expansion in order to supply their consolidated portfolio, and this reduces the potential market share for new entrants. A capacity mechanism in the form of a decentralized auction allows vertically integrated incumbents to favour their subsidiaries. In order to avoid this scenario and reduce entry barriers for new entrants, a transparent and centralized auction should be launched for the coverage of the entire reliability demand, in which the perfect match between the vertically integrated generator’s supply and retailer’s demand is not possible. This is the case in the UK, where a capacity mechanism based on centralized auctions is currently being implemented.23 However, the capacity mechanism

Agency) was implemented to contract large power projects, mainly aiming at the untapped hydropower potential present in the country, which is completely managed by the Peruvian state. 20 21 Batlle and Rodilla (n 3). Batlle and Rodilla (n 3). 22 OFGEM, The retail market review—findings and initial proposals, Consultation paper, March 2011. 23 DECC, Implementing EMR (n 8).

7.3 Target market

125

currently under design in France, another market characterized by vertical integration between generation and retailing, will be a decentralized one. The recently released guidelines from RTE24 impose capacity obligations on retailers, who then need to procure capacity certificates via bilateral contracts in order to meet these obligations. Apparently, no auction scheme is foreseen, not even a decentralized one, and exchanges are expected to take place through direct negotiations. This design highlights the previously mentioned concerns regarding vertical integration and permits incumbents to create entry barriers for new entrants. Regarding the allocation of responsibility in estimating future demand, some authors25 have expressed their preference for a decentralized approach, suggesting that distribution companies should be in charge of forecasting future demand, with an appropriate scheme of penalties for over/under estimating. This methodology is based on the assumption that distributors have a clearer vision of the extent of the demand, and could better estimate its growth. While this argument cannot be disproved, it should be clear that the ultimate responsibility for security of supply is always with the regulator (or the system operator often acting on its behalf ). Moreover, it is important to remember that one of the key reasons for the implementation of capacity mechanisms is the inability or unwillingness of the demand side to properly hedge in the very long term (all demand side to some extent, but most relevantly residential consumers). If the design of efficient penalties imposed on generating units for noncompliance is in any case a complex or controversial issue, setting the right incentives and credible penalties for retailers to properly estimate their long-term demand, is an even more troublesome task. Accordingly, demand forecasting should be made by the regulator itself.

7.3.2 The selling side The regulator, or the agent running the auction, has to decide who is allowed to present bids. Two major decisions have to be made: first, whether both existing and new investments should be allowed to participate; and second, if technological priorities should be explicitly applied.

7.3.2.1 The role of existing plants The regulatory decision to implement an auction mechanism in order to bring in new generation could be perceived as a market intervention that negatively affects existing plants, as this new generation depresses prices in the spot market. From a rigorous economic theory perspective, it can be argued that existing power plants should also be allowed to compete on equal terms in order to receive the same price. The equal

24

RTE (n 7). Rodrigo Moreno, Luiz A. Barroso, Hugh Rudnick, Sebastian Mocarquer, and Bernardo Bezerra, ‘Auction approaches of long-term contracts to ensure generation investment in electricity markets: Lessons from the Brazilian and Chilean experiences’ (2010) Energy Policy 38 (10), 5758–69. 25

126

The System Adequacy Problem

treatment of existing and new power plants would guarantee, in the long term, an efficient signal to drive generation expansion. In fact, this approach is common in most markets.26 However, the previously mentioned reasoning is often refuted with two objections. First, political interests often lead to the simple conclusion that, since their fixed costs are sunk, there is no need to allow existing generation units to participate in the auction or to receive the resulting price. Excluding these units from the auction certainly minimizes supply costs for consumers in the short term, but, as stated, it is far from being clear that this decision will not result in higher supply costs in the long run, since investors might conclude that they will not be able to capture the long-term marginal price when their plants will be considered as existing. Secondly, the vast majority of the existing generating units were installed under the former traditional cost-of-service scheme, which guaranteed the recovery of the capital costs. This, and the fact that they may be publicly owned or controlled, provides some justification for the regulator’s decision to discriminate between existing and new generation units, especially in the South American context. However, even if this regulatory solution is chosen, it would be important that existing plants, which have fully recovered investment costs, would also be given an economic signal encouraging them to manage their units in order to enhance their availability under scarcity conditions. This is the underlying motivation behind the proposal to define two different reliability-oriented payments: (a) an adequacy-oriented payment available only for new entrants and (b) a firmness-oriented payment, targeting all the generation units in the system (both new and existing).27 Therefore, the most common distinction for this design element is between mechanisms where different auctions are organized for new and existing plants, and schemes that mix the two categories in the same auction. Brazil has implemented separate auctions from the beginning, while in US-ISOs, Chile, Colombia, and Peru existing and new plants participate in the same tenders. However, there is then discrimination in Colombia and ISO-NE between new and existing plants, based on the duration of the contract, which is one year for existing plants and up to twenty years (Colombia) or five years (ISO-NE) for new projects. Another approach is to allow existing generation units to take part in the auction only as price-takers (ie only allowing them to bid at zero or close to it), while leaving the task of setting the auction price to the new power projects (which act as price-makers). This is the current design in New England’s Forward Capacity Market,28 while in Colombia a system of different price caps to be applied in special auction conditions is used to further discriminate between existing and new generation units.

26 For example, as is the case in most markets, in the real estate market, it is unquestioned that old and new houses compete at the same level. 27 Developed in Carlos Batlle, Carlos Solé, and Michel Rivier, ‘A new security of supply mechanism for the Iberian Market’ (2008) The Electricity Journal 21 (2) 63–73. 28 Apparently, this is also the design chosen for the UK capacity auction, in which existing plants will be allowed to participate only as price-takers. DECC, Implementing EMR (n 8).

7.4 Lag period (or lead time)

127

7.3.2.2 Technological neutrality or not? The second issue related to this design element is the introduction of technologyspecific auctions, which have been launched by the regulators of several Latin American countries. This was done either to promote renewable energy projects,29 or to foster the exploitation of the continent’s large untapped hydropower resources.30 From a theoretical point of view, in liberalized power systems, decisions about capacity expansion, particularly the choice of technology, are left to the market agents, who are expected to bear the related risk of a poor assessment of the system’s future needs. This approach forms the basis of the original decision to liberalize electricity markets, and it is thought to allow the most economically efficient combination of technologies to enter the system. This kind of intervention, in the form of technology-specific auctions, is usually accepted if it aims at the very long-term strategic objectives of the country, which would not otherwise be achieved because they are not considered by market agents when making their decisions. In this case, these long-term objectives should be specified at the moment of launching the technology-specific auction, and subsequent regulation should be consistent with them. However, it is important to bear in mind, that when there is a broad range of potential technologies, certain auction details (eg contract provisions) often might lead to de facto discrimination between different generation sources. For example, defining a lag period (ie the maximum time available for building a power plant) of three years makes it almost impossible for large hydro plants to compete against conventional thermal generation. This is discussed in more detail in the following section.

7.4 Lag period (or lead time) The lag period is the time between the contract signature and the date when the contract comes into force. Unless they are involved in previous contractual commitments, existing plants need no lag period, because from a technical point of view they can start producing electricity immediately. However, for administrative reasons, contracts with existing plants usually consider a lag period between a few months and one year. On the other hand, for new generation projects the lag period represents the maximum time available for construction. Obviously this parameter heavily conditions the competitiveness of the different plants and technologies in the auction. In the case of South America, a critical look at the length of the lag periods clearly illustrates this point.

29 For example, the case of the RER auctions in Peru or the Brazilian auctions to add biomass or wind generation, see for example Gabriel Cunha, Luiz A. Barroso, Fernando Porrua, and Bernardo Bezerra, ‘Fostering wind power through auctions: the Brazilian experience’, IAEE Energy Forum, Spring 2012, pp 25–8; or Paolo Mastropietro, Carlos Batlle, Luiz A. Barroso, and Pablo Rodilla, ‘Electricity auctions in South America: towards convergence of system adequacy and RES-E support’ (2014) Renewable and Sustainable Energy Reviews 40, 375–85. 30 For instance, the Proinversión auctions in Peru (n 19) or the project-specific hydropower auctions in Brazil.

128

The System Adequacy Problem

In Brazil, three different kinds of auctions are implemented with the main difference between them being precisely the lag period. The so-called A1 auction (one-year lag period) is meant for existing generating units, while the A3 auction (three-year) clearly aims at adding thermal generation and A5 is intended for large hydro projects. In Peru, the first design defined a three-year lag period, clearly in line with the governmental desire to exploit the Camisea gas pipeline constructed in the late nineties through the installation of gas plants. A few years later, the government considered the need for new large hydro investments, and questioned the auction’s design. This concern led to an out-of-the-electricity-regulation call for tenders aimed specifically at attracting this generation technology.31 The Colombian reliability charge mechanism aims to attract both thermal and hydro plants. Colombia introduced two related but separated auctions: (a) a Firm Energy Obligation (OEF) auction32 with a lag period of four and a half years, more suitable for and implicitly targeting, thermal plants, and (b) an auction with a lag period of seven years, focusing on new hydropower projects.33 It appears that the only way to design an auction in which more than one technology could equally compete would be to allow for different lag periods. Therefore, if no strategic objective results in a preference towards certain technologies, the only approach to avoid this de facto technology discrimination would be to allow all kind of plants to bid in the auction with the assurance that they have enough lead time for installation. However, this leads to another relevant problem, as it is very difficult to define proper criteria to transparently compare different bids. For instance, it would be difficult to compare a plant bidding for a short lag period and a higher price with a plant bidding for a longer lag period and a lower price. Therefore, the regulator faces a trade-off in defining whether single or multiple lag periods are to be considered. In the United States, the most common approach has been that of considering one single lag period, usually of three years (PJM and ISONE). However, it is worth noting that these mechanisms principally target thermal plants (mainly CCGTs, which have short construction times) and demand response, which do not need long lag periods.

7.5 Contract duration The duration of the contract offered in the tender is one of the most relevant design elements of a long-term electricity auction. As mentioned throughout this chapter, the main objective of electricity auctions is to hedge the generators’ risk against the high volatility of spot market prices and more importantly, against regulatory risk, in order 31 In 2002 and 2003, Decree No 027/2002/PCM and Decree No 095/2003/EF created and renamed, respectively, the agency finally known as Proinversión (The Private Investment Promotion Agency). This institution is in charge of fostering private investment in strategic infrastructure development, hydropower plants, and other large power projects within this scope. 32 OEF stands for Obligación de Energía Firme (Firm Energy Obligation). 33 The so-called GPPS auction is organized in Colombia after the OEF auction and it is designed for those plants with construction times exceeding the OEF lag period. The reserve price in the GPPS auction is the price cleared in the OEF auction.

7.5 Contract duration

129

to facilitate their access to financing and to improve the overall attractiveness of the investment. Keeping this in mind, the contract duration should be long enough to provide investors in new generation projects with the necessary stability for investment. The long-term incentive to attract new investments is enclosed in this parameter and its determination has a dramatic impact on the results of the auction. As in the case of the lag period, the contract duration may vary within an auction, and it is also possible to differentiate between plants (new or existing) and technologies (eg hydro or thermal) in the same action. The Brazilian experience clearly illustrates how tenders can be ‘guided’ through their design elements. The A1 contracts, since they are aimed only at hedging market prices according to the El Niño Southern oscillation,34 have a duration of five to eight years. Conversely, the length of the contracts offered in the A5 auctions ranges from fifteen to thirty years, confirming our previous statement that these auctions are clearly designed to attract large hydro projects. Similar conclusions can be put forward when reviewing the Peruvian auctions, where, in the Proinversión auctions,35 a contract duration can extend beyond ten years. Regarding new plants, the contract duration should reflect the capital intensity of the technology. Obviously, thermal plants need shorter terms to mitigate their risk than hydropower plants. In any case, at the point of determining the contract duration, consideration must be given to how the project is financed and the discount rate involved. In fact, due to the effect of the discount rate, the more distant future income is from the present, the lesser is its impact on the net present value. Cash-flows perceived thirty years from now are almost irrelevant in the decision making when high discount rates are considered. Therefore, contract durations above this value are seldom justified with the high discount rates (higher investment risk) used in the generation sector in South America. On the other hand, in North America, common contract durations are significantly lower. In the ISO-NE Forward Capacity Market, existing plants are entitled to only a one-year contract, while new plants can choose contract durations from one to five years. This is probably due, once again, to the different technological targets (thermal plants, in this case), but also to the lower country risk. Moving beyond the regional scope of this chapter, the capacity mechanism currently discussed in Italy is based on reliability option contracts to be procured in a centralized auction with no discrimination between new and existing plants.36 These contracts, as proposed by the Italian energy regulator (AEEG)37 have a lag period of at least four years, thus clearly targeting new generation projects. Nevertheless, the contract durations foreseen are equal to three years, which could even be reduced to one year in some cases. While these contract durations could be acceptable for existing generation, they are clearly not sufficient to hedge investors’ risk and, therefore, they are not suitable for attracting new plants. This type of design could hamper the effectiveness of 34 El Niño Southern Oscillation, or ENSO, is a naturally occurring phenomenon that involves fluctuating ocean temperatures in the equatorial Pacific. They can be anomalously warm or cold for long periods of time causing variations in regional climate patterns. 35 36 37 See n 18. See chapter 17. AEEG (n 6).

130

The System Adequacy Problem

the mechanism in guaranteeing system adequacy. Actually, in the current context of overcapacity that the Italian system is experiencing, it seems that the objective of a capacity mechanism with such short contract durations is to remunerate the fixed costs of the existing generation mix, which cannot be recovered in the energy market due to prices being lower than expected, rather than to foster new investments.

7.6 Defining the requirements associated with the ‘reliability product’ There are many types of capacity or energy products that the regulator can purchase on behalf of demand to solve the security of supply problem (also known as reliability products). Among others, these are capacity credits/obligations, firm energy contracts, reliability contracts/options, strategic reserves, etc. Today, the use of these terms is still vague and can be misleading, as it is not clear which actual characteristics and commitments they are supposed to define. In general, the reliability product can be more properly defined and classified based on the following questions: (a) Is the product energy-based or capacity-based? (b) Does the contract imply a physical commitment, financial or both? (c) When are agents selected in the auction required to fulfil their commitment? (d) How are they penalized in case of underperformance? (e) Is there any regulatory limitation on the quantity of product that each agent in the power system can sell via the mechanism? Dealing with these questions, in this section, permits the identification of different kinds of reliability product.

7.6.1 Reliability in capacity- and energy-constrained systems There is a relationship between the security of supply problem that is to be addressed and the product the regulator should define. This first calls for properly identifying the system needs (energy or capacity). Not doing so is frequently a source of policy failure in the design of the mechanism. In this line, it is necessary to understand which kind of scarcity conditions can be expected in the system. The main classification that can be applied divides systems into those that are capacity-constrained and those that are energy-constrained. In capacity-constrained systems, scarcity problems arise because there is not enough installed capacity available (MW) to satisfy demand at a given moment due to, for instance, the forced outage of thermal plants and/or minimum wind output. Aggregating all the hours, the system could certainly have enough energy available to satisfy demand on a given day (more than enough production capacity in off-peak hours), but it lacks installed capacity to satisfy peak demand. This type of potential scarcity conditions are usually found in the European and North American systems, which has prompted operators of these systems, and the market participants, to be primarily concerned with modelling the very short term in great detail. Note that in these systems a high penetration of non-dispatchable wind and solar generation ensure that enough energy is there through the year, but capacity is not necessarily available at the time it is needed the most.

7.6 Defining requirements associated with ‘reliability product’

131

The situation is quite the opposite in the energy-constrained systems, which could certainly satisfy peak demand, but would not be able to supply the demand during the remaining hours of the day/week. Thus, rationing takes place due to a lack of available energy, not capacity. A large proportion of Latin American systems traditionally fall into this category due to the large share of hydropower in their generation mixes. The availability of reservoirs with a large storage capacity has historically reduced the necessity of considering the short-term operation in any detail. However, in these systems it has been of critical importance to suitably represent uncertainty and to determine the optimal management strategies for hydro resources and their medium-, long, and very long-term interaction with thermal facilities. This distinction must obviously be reflected in the design of the reliability product. On the one hand, a capacity-constrained system has to remunerate the agents for their ability to provide instantaneous power to cover peak demand. On the other hand, an energy-constrained system should reward the capacity of agents to manage their resources in order to guarantee their availability in those periods when energy scarcity occurs, for example, during a dry year. This consideration is confirmed by the diversity of capacity mechanisms introduced in different systems. In the UK, a system with a very large share of conventional thermal plants, the currently implemented capacity mechanism is completely focused on the capacity that agents can inject into the network. By contrast, the energy-constrained Brazilian system ensures the security of supply through long-term capacity auctions, in which generators offer full-energy contracts, with yearly settlements. An intermediate solution between these two extremes can be found in the Colombian energy-constrained system. The capacity mechanism implemented in this country (Firm Energy Obligations scheme) requires agents to deliver the energy committed in the auction during those days when the spot price exceeds the strike price at least once.

7.6.2 When are agents selected in the auction required to fulfil their commitment? The definition of the critical period Certain types of capacity mechanisms define a critical period (also called scarcity conditions or near-rationing conditions), during which each agent with a reliability commitment must deliver the product sold in the auction. For the countries analysed in this chapter, this does not always apply. In Brazil, Chile, and Peru, where standard future supply contracts are auctioned, the generators selected by the tender mechanism have to deliver electricity according to the contracts they sign. No critical period is defined, and even if the system has some method to identify near-rationing conditions, this does not affect the contract provisions. The Colombian firm energy obligation (OEF) mechanism is, in contrast, based on a definition of scarcity conditions. The spot market price is used as a critical period indicator and the scarcity conditions are defined as the period of time during which the spot market price exceeds a predetermined strike price.38 The Colombian mechanism is based on option contracts whereby 38 Carlos Vázquez, Michel Rivier, and Ignacio J. Pérez-Arriaga, ‘A market approach to long-term security of supply’ (2002) IEEE Transactions on Power Systems 17 (2), 349–57. Actually, the Colombian

132

The System Adequacy Problem

the seller, in exchange for a fee, commits to provide the buyer with electricity not at the actual spot price, but at the strike price. Therefore, the strike price has two functions: on the one hand, it identifies the critical period, on the other hand, it acts as a soft price cap (only for the generation awarded with a reliability commitment in the auction). In Colombia, the strike price calculation is linked to a basket of fuel price indexes updated through the Platts US Gulf Coast Residual Fuel No 6 1.0 per cent sulfur fuel oil.39 A number of general recommendations on the selection of the critical period indicator were already provided by Batlle and Pérez-Arriaga,40 who identified the short-term market price as the best ‘thermometer’ of scarcity conditions in a market environment. This consideration should become increasingly valid in the future, with greater elasticity of demand. In fact, as long as the share of completely inelastic demand in the market (ie the demand that bids at the price cap) decreases, it will become increasingly difficult to define the demand which must be served and, consequently, to identify near-rationing conditions based only on the comparison of peak demand and available generation. This critical period indicator obviously assumes the presence of a liquid short-term market in the system (reference market). The selection of the reference market also affects the type of scarcity conditions that are covered, and those that the regulator wants to be covered, by the capacity mechanism. On the one hand, day-ahead markets are only capable of capturing emergency situations related to the combination of high loads (as peak winter demand) and reduced supply (due to fuel constraints or a dry year that limits hydro production), that is, pure generation adequacy issues. A capacity mechanism using a day-ahead market as a reference would thus address this type of scarcity. On the other hand, intraday and balancing markets are also subject to price fluctuations due to more or less sudden events, such as the outage of a nuclear plant or, in those systems with high renewable penetration, the fall of intermittent generation due to bad forecasting. These are events which provoke temporary generation scarcity even if the load is far from its peak, and have a time horizon longer than the one covered by ancillary services, ie flexibility issues. Therefore, by adding intraday and/or balancing markets as reference markets (there can be more than one reference market), a capacity mechanism may address not only adequacy but also flexibility issues. However, the selection of the relevant market also has to take into account the signal that the capacity mechanism is providing to the generation mix. While all units are more or less technically capable of producing, if notified one day ahead, certain technologies (base-load technologies, such as coal power plants) would not be able to take part in the balancing market, because they cannot respond in such a short time frame, due to ramp constraints.41 Therefore, a capacity mechanism using the balancing market as the reference market provides mechanism, differently from what is suggested by Vázquez and others, considers a daily obligation. Also, if the spot price exceeds the strike only for one hour, scarcity conditions are applied to the entire day. 39 Comisión de Regulación de Energía y Gas, Res No 071/2006, Resolución por la cual se adopta la metodología para la remuneración del Cargo por Confiabilidad en el Mercado Mayorista de Energía, 3 October 2006. See Annex 1, para 1.4. 40 As specified by Carlos Batlle and Ignacio J. Pérez-Arriaga, ‘Design criteria for implementing a capacity mechanism in deregulated electricity markets’ (2008) Utility Policy 16 (3) 184–93. 41 See the explanation of ramp constraints in chapter 3 at n 4.

7.6 Defining requirements associated with ‘reliability product’

133

agents with a signal that discourages the installation of new base-load units, and this may not be the regulator’s objective. Where the short-term market price is selected as a critical period indicator, and a strike price has to be determined in order to signal scarcity conditions, two aspects must be taken into consideration.42 First, the strike price must be predetermined by the regulator based on a formula available to all interested agents and must be unique for all those taking part in the auction. Allowing agents to bid on both the strike price and the option fee associated with it, would add a significant level of complexity to the auction mechanism. Since the bids would not be simple price-quantity pairs, their comparison would be possible only through a model simulating the market behaviour. This would reduce transparency and add a degree of arbitrariness to the process. Secondly, the strike price must be determined in such a way that it does not interfere with the normal functioning of the short-term market. This can be achieved by fixing it well above the SRMC43 of generation production (and demand response measures). In any case, the selection of the critical period indicator should take account of long to very long-term contracts signed under the framework of capacity mechanisms. A critical period indicator that may appear appropriate in the present could become unsuitable in the future because of changes in the generation mix or fuel prices. Going beyond the regional scope of this chapter, the recently proposed capacity mechanism in France44 apparently identifies scarcity conditions using the temperature as the critical period indicator. However, a critical period indicator based on temperature does not take long-term contracts into account as the correlation between scarcity conditions and temperature may change in the future (eg due to a large penetration of RES generation), and there is a risk that long-term contracts are no longer providing the expected reliability. The critical period indicator used in the UK capacity mechanism is also not aligned with the recommendations presented here. In fact, in the UK capacity mechanism design, the scarcity conditions are replaced by the concept of ‘system stress’, defined as ‘any settlement periods in which either voltage control or controlled load shedding are experienced at any point on the system for 15 minutes or longer’ and communicated to agents with a reliability commitment at least four hours in advance through a ‘capacity market warning’.45 As a result, the scarcity conditions are no longer linked to the reserve margin and they could also occur when the demand is, for example, 70 per cent of peak demand. In this case, each reliability provider ‘will only be required to be generating electricity or reducing demand up to 70% of their full capacity obligation.’46 This approach has two clear drawbacks. First, the determination of system stress is somehow arbitrary and completely unpredictable. This uncertainty will be taken into consideration by agents, when assessing the impact that taking part in the capacity mechanism can have on their revenues, which will be reflected in their bids. Secondly, it hampers DSR participation in the capacity mechanism, as DSR is only available to provide a certain demand reduction during peak demand conditions.

42 44 46

43 According to Carlos Vázquez et al (n 38). SRMC stands for short run marginal costs. 45 RTE (n 7). See chapter 14. DECC, Implementing EMR (n 8) p 107. See chapter 22. DECC, Implementing EMR (n 8) p 108.

134

The System Adequacy Problem

7.6.3 Penalties for non-compliance The non-fulfilment of contract commitments must be penalized by the regulator. The penalty should be high enough to dissuade the selected bidders from failing to accomplish their obligations. At the same time, the penalty should not be excessive in case of prolonged technical unavailability. This could be achieved through an hourly penalty associated with a correction factor diminishing with the duration of the unavailability, or through the use of cumulative penalty caps. In the latter case, ‘soft caps’ should be used, allowing agents who have reached the penalty cap to reduce their penalty through performances above their target.47 This ensures that these agents continue to have an incentive to fulfil their contract once they have reached the penalty cap. Another distinction to be made is between implicit and explicit penalties for underperformance. Without entering into the details of the different possible capacity mechanisms, when an agent does not fulfil its commitment, it can be penalized in two ways. First, it could be required, in case of ‘non-delivery’ of the reliability product, to procure an equivalent amount of the reliability product in the electricity market, in order to honour its contract, even if not through its own assets (implicit penalty). In addition, when designing the capacity mechanism, the regulator can introduce an explicit penalty, in the form of an extra fee for non-compliance, calculated according to a predetermined formula (explicit penalty). Vázquez et al, in presenting the reliability option principles, asserted the need for an explicit penalty in order to discourage agents without reliable generation capacity from participating in the auction.48 In South American mechanisms, the only penalty usually applied is the implicit one. The Colombian firm energy obligations, for example, do not include any explicit penalty. According to several experts, this exposes a potential weakness in the Colombian capacity mechanism. This, for example, is the view of the Colombian Wholesale Electricity Market Monitoring Committee49 after scarcity conditions occurred in 2009–2010. In its report, it is stated that hydro generators prefer not to respect their reliability option contract in the future than to face a certain economic loss at present. This behaviour is clearly related to the lack of an explicit penalty, which would increase the magnitude of the effect of future non-compliances, consequently increasing its influence upon present decisions. Conversely, in the North American context it is possible to find explicit penalties in the case of underperformance during scarcity conditions. In the Forward Capacity Market of ISO-NE, a ‘shortage event’ is a shortage of electricity operating reserves, defined as any period of thirty or more consecutive minutes of system-wide deficiency in operating reserves. Agents who have assumed the obligations of capacity supply receive payments for the ability to produce electricity when required. During shortage events, their performance is measured, and those that are unavailable to produce 47 It must be underlined that a soft penalty cap can be applied only in those schemes where the agents can actually over-perform with respect to their commitment. This could happen because of a production that exceeds the maximum quantity that the agent was allowed to trade in the capacity mechanism, or due to the specifications of the product to be provided during scarcity conditions. 48 Carlos Vázquez et al (n 38). 49 CSMEM, Comité de Seguimiento del Mercado Mayorista de Energía Eléctrica, Informe No 53 – 2010. Experiencias de la intervención del MEM bajo efecto del Niño 2009–10. Report released on 13 October 2010.

7.6 Defining requirements associated with ‘reliability product’

135

electricity are penalized with reduced capacity payments. More precisely, every time a shortage event occurs, the system operator calculates the so-called Shortage Event Availability Score for each agent holding a capacity supply obligation. Based on this score, a Shortage Event Availability Penalty is defined as: (Annualized Forward Capacity Payment)  PF  (100% Score)

Shortage Event Availability

where PF (Shortage Event Penalty Factor) is 5 per cent for events lasting five hours or less and increased by 1 per cent for each hour above five hours. If the Shortage Event Availability Score is 100 per cent (ie capacity supply obligation fully fulfilled), the agent is not penalized; otherwise the explicit penalty is applied. These penalties are subject to a number of daily, monthly, and annual caps, in order not to exceed the Annualized Forward Capacity Payment.50

7.6.4 Constraints on tradable quantities Theoretically, there should be no need to set constraints on tradable quantities in the capacity mechanism, provided that (a) the explicit penalties for underperformance are properly designed, in such a way as to favour more reliable generation technologies, and (b) collaterals are high enough, providing the system with a financial guarantee in case agents’ estimations prove to be wrong. Nevertheless, most regulators and system operators in systems with capacity mechanisms, have developed a methodology for setting an upper and a lower limit on the quantities tradable by each agent. Introducing constraints on tradable quantities is based on the following reasoning. Without an upper limit, the number of cases where agents overestimate the quantity of reliable product they are capable of producing in scarcity conditions might compromise the financial stability of the entire power sector, with potential repercussions for security of supply. In order to avoid this, very large and costly warranties should be introduced. However, this could dramatically reduce participation in the capacity mechanism. Without a lower limit, some agents could be tempted to behave strategically, withholding part of their capacity from the auction, with the objective of increasing the clearing price. Therefore, setting a minimum amount that the agents are required to trade would help to manage market power issues. These methodologies resulted in the development of the concepts of firm energy or firm capacity,51 and establishment of some sort of prequalification phase, during which these parameters are calculated for all agents willing to participate in the auction. In South American auctions, most of the regulators have introduced such methodologies, Chile being the main exception.52 A detailed description of these calculation methods 50

At the moment of writing, a new penalty scheme is being approved, see FERC, Order on Tariff Filing and Instituting Section 206 Proceeding, docket no. ER14-1050-000, issued on 30 May 2014. This reform, based on the Pay-for-Performance principle, reinforces the current scheme, with explicit penalties that will be raised up to 5,455 $/MWh. 51 Depending on the specific design of the reliable product auctioned. 52 Theoretically, the contracts signed in Chilean auctions are not required to be covered by any firm energy certificate and the distribution companies have to assess the bidders’ credibility on their own. However, generators have to specify to the regulator, on a yearly basis, which plants will be used to cover the contracted demand.

136

The System Adequacy Problem

exceeds the scope of this chapter, but by way of example, we explain, in the following, how this is done in the Colombian case for South America and in ISO-NE for North America. In Colombia, generators willing to take part in the OEF auction must be backed by firm energy certificates (also called ENFICC), whose calculation methodology has been developed by CREG, the Colombian energy regulator.53 The ENFICC of hydraulic plants is calculated using a computational model (HIDENFICC), which determines the maximum production that can be obtained monthly from a hydro plant during dry periods. The lower ENFICC limit that a generator or investor can trade is termed ‘ENFICC Base’ and corresponds to the minimum energy obtained by the maximization model. The upper ENFICC limit corresponds to the energy that a generator can produce with a probability of 95 per cent, called ‘ENFICC 95%’. If a generator or investor is willing to trade an ENFICC higher than the ENFICC Base in the auction, without exceeding the ENFICC 95 per cent, it should back this difference with a financial warranty. On the contrary, the ENFICC of a thermal plant is calculated based on the plant’s generation capacity, fuel availability, the number of generating hours per year, and an index that incorporates the historical restrictions imposed upon the plant, which limits its maximum energy generation. As regards renewable energy, CREG has recently outlined the methodology for calculating the ENFICC Base and the ENFICC 95 per cent for wind farms.54 The Forward Capacity Market of ISO-NE follows a different approach, based on historical availability. Existing plants have recognized summer and winter qualified capacities, calculated as the median of the claimed capability ratings for the five most recent summer and winter seasons. For hydro plants with daily cycles, the qualified capacities are based on the seasonal average, calculated using the 50th percentile flow rate. As regards new plants, project sponsors must submit an expression of interest to the system operator which contains the requested summer and winter qualified capacities. Based on the financial reliability of the project, the construction schedule, and the requested qualified capacities, the expression of interest can be accepted or rejected. The qualified capacities are then adjusted during the following auctions based on real data from the plant. Finally, it must be underlined that firm energy and firm capacity are the basis on which plants and projects are remunerated in the capacity mechanism. Therefore, a revision procedure that punishes underperformance by administratively reducing the firm energy or capacity of a plant can constitute a very effective penalty scheme.

7.7 Indexation and warranties There are several other design elements that, although often considered as secondary, can have a significant impact on the final results of an auction. In this section three are singled out for review. 53

Comisión de Regulación de Energía y Gas (n 39). Comisión de Regulación de Energía y Gas, Resolución No 148 de 2011, por la cual se define la metodología para determinar la energía firme de plantas eólicas, 21 October 2011. 54

7.7 Indexation and warranties

137

7.7.1 Indexation formulas In the South American experience, in line with what has been the traditional method of remunerating utilities and independent power producers in the previous regime, all the economic parameters set by the contract are usually subject to indexation formulas that determine their future evolution. Commonly, these formulas, besides being linked to the retail price index (in the US) or the exchange rate of the dollar with the local currency (in South America), use the international price of fuels as a reference, with a view to determining how this parameter will affect the operating costs of power plants. However, economic theory recommends assigning risks to the agents who can best manage them. Indexing the energy price to fuel prices implies allocating this risk to electricity consumers who have no ability whatsoever to properly manage it. Thus, it would be better to avoid fuel indexation in the contracts, since, in principle, generators should be able to hedge this risk more efficiently by signing contracts on the international markets for commodities. Unfortunately, it is also true that these markets present a low liquidity in the long to very long term (ie longer than five years). Therefore, the optimal solution would be to design an incremental ‘indexation weight’, ie the percentage dependence of the contract prices with respect to international fuel prices. Another question, concerning indexation, is whether to use a single formula to index all the contracts or to allow agents to include the required indexation within their bid. The latter approach, used for example in Chile,55 creates a challenge when comparing different bids, because their competitiveness in the long term varies broadly according to this parameter. A general recommendation is to define a unique indexation formula for all the generators involved in the auction in order to increase transparency and keep the auction format as simple as possible.

7.7.2 Financial warranties With respect to warranties, as in other markets, bidders selected through the auction process should provide a monetary endorsement to cover at least part of the potential penalties for non-fulfilment. In order not to introduce large warranties that could limit participation in the auction, this endorsement could be achieved by retaining part of the first contract payments until a targeted figure is reached. However, for both penalties and warranties, it must be stressed that their determination is strictly related to the selection of the reliability product and, in particular, to the constraints on the tradable quantities, as mentioned above in section 7.6.4. In fact, the higher the penalties for non-compliance and the higher the warranties required, the lower the need to be

55 As mentioned by Maurer and Barroso (n 13) in Chile, the indexation formulas are determined and published by the regulator in the form of a multivariable linear function of fuel and inflation indices in which each multiplying factor is ultimately adjusted by each bidder, thus creating different indexation requirements. However, indexation formulas are not taken into account by the auctioneer during the allocation process, thus affecting the overall economic efficiency of the process, since the set of winners could dramatically change if indexation formulas were incorporated into the clearing mechanism.

138

The System Adequacy Problem

strict on defining the maximum value each unit can trade at or the level at which it can be remunerated.56

7.8 Summary and high-level recommendations Long-term auctions have been selected by many regulators on the American continent as the most effective instrument to guarantee the adequacy of the system. They reduce the risk associated with the long-term volatility of spot market prices and the potential regulatory interventions by guaranteeing and fixing ex-ante part of generators’ future income through an, in principle, efficient (if competitive) and transparent capacity mechanism. This economic signal facilitates project financing and fosters the installation of new generation capacity. However, these instruments must be designed carefully. In this chapter, we have identified, reviewed, and discussed the key design elements of long-term capacity auctions, extracting guidelines from the South and North American experiences. The main lessons learned can be briefly summarized as follows. On the buying side, the best option seems to be to involve all the system demand in the auction, since the entire system demand benefits from the increased system reliability, thus avoiding free-riding issues. The demand which is not willing to be covered by the reliability mechanism, can then offer demand response bids, or even energy efficiency bids, in the auction. However, this option should only be open to users who can actually be disconnected during scarcity events, and after a careful determination of the customer baseline. As regards the centralization level, the most efficient solution appears to be a centralized auction that covers the whole system demand. A centralized auction has a threefold benefit. First, it allows the exploitation of economies of scale in generation. Secondly, it mitigates the impact of vertical integration of electricity companies, which is still an issue in several power sectors, and gives an opportunity to new entrants. Last but not least, it enhances transparency and guarantees an equal ‘system-adequacy’ price for all consumers. On the selling side, existing and new generators can compete in the same auction, as long as specific measures are taken in order to differentiate between the two categories, such as defining existing generators as price-takers in the auction or setting different price caps. As regards technology-specific tenders and the determination of such parameters like, for example, the lag period for new power plants, which can implicitly include or exclude certain technologies, it must be understood that a certain degree of discrimination will always be present. A capacity mechanism which accurately defines the reliability product and the associated parameters, will always favour a certain technology. However, the alternative approach, ie letting bidders specify the parameters of the reliability product (lag period, contract duration, strike price, etc), leads to the undesirable result of having to ‘compare apples and oranges’ in order to clear the auction. 56

Batlle and Pérez-Arriaga (n 40).

7.8 Summary and high-level recommendations

139

Regarding the critical period indicator, the best choice is the short-term price of the reference market in the system, which is the most suitable measure of scarcity conditions in a market environment. Having the short-term price as the critical period indicator obviously assumes the presence of a liquid power exchange in the system, but this, in fact, is now considered as an essential feature of any efficient wholesale market. Therefore, in those systems where such a reference market is not yet in place, the implementation of a capacity mechanism of this sort could be an opportunity to foster the development of a liquid short-term market (day-ahead or balancing). Finally, as regards penalties, it is essential to define a robust explicit scheme which adds an extra fee on top of the implicit penalty of having to purchase, in the market, the electricity necessary to fulfil the commitment if the agent is not able to produce it with its own assets. An explicit penalty discourages those bidders that are not backed by reliable generation capacity and encourages those selected in the auction process to manage their units in such a way as to improve their availability in scarcity conditions, thus enhancing the firmness of the generation mix

8 The Generation Mix, Price Caps, and Capacity Payments Bert Willems

8.1 Introduction This chapter provides an economic analysis of capacity mechanisms in electricity markets. It considers a single market distortion in the form of a price cap on the spot market, which limits the price that generators receive, and studies how this distortion affects investment incentives for different generation technologies. It then discusses how a capacity mechanism can be used to restore the efficient generation mix. The goal of this chapter is to provide a simplified analysis of generation investments and highlight some key economic concepts. I show, for instance, that although a price cap affects investment levels in peak-load generation and not in base-load capacity, capacity mechanisms should be technology-neutral, and provide the same incentives to base-load and peak-load generators. Furthermore, I show that the introduction of renewable energy reduces the profitability of existing generators, but that this does not imply that capacity mechanisms should be used to bail-out generation plants. The chapter’s framework relies on a decades-old pre-liberalization deterministic graphical analysis of optimal generation portfolios, using load-duration graphs and total cost functions.1 The methodology is often used as an educative tool. Here we add a price cap and capacity payments.2 In this chapter, I shall not discuss whether or not capacity mechanisms should be part of a well-designed power market. One could think of several market failures (or regulatory failures) that distort investment incentives and for which a capacity mechanism is only one of the many possible remedies. For instance, a price cap might be necessary to address the market power of dominant generators,3 as energy prices might

1

This method relies on a very simple presentation of the electricity market, as discussed in section 8.2. It assumes there will be inelastic and predictable demand fluctuations and very flexible production plants. Without those assumptions, a graphical solution is no longer possible, and simulations or mathematical analyses become necessary. For the use of this method see, for instance, Sally Hunt, Making Competition Work in Electricity (New York: John Wiley & Sons, 2002), Appendix E; and Allen J. Wood and Bruce F. Wollenberg, Power Generation, Operation, and Control, 2nd edn (New York: John Wiley & Sons, 1996), Chapter 8.2 on load duration curves. 2 See for instance Gonzalo Sáenz de Miera, Pablo del Río González, and Ignacio Vizcaíno, ‘Analysing the impact of renewable electricity support schemes on power prices: the case of wind electricity in Spain’ (2008) Energy Policy 36 (9), 3345–59; and Paul L. Joskow, ‘Capacity payments in imperfect electricity markets: need and design’ (2008) Utilities Policy 16 (3), 159–70. 3 Market power is a deviation from perfect competition and therefore a market failure in the strict sense. Whether this market failure justifies additional regulation depends on the extent of the failure and its impact on welfare. The use of a price cap itself might constitute a form of regulatory failure, if there are

8.2 Optimal generation mix

141

be artificially low because of the monopsony power of the network operator, financial markets might be insufficiently liquid for firms to hedge price and volume risk, regulators might have to determine the value of undelivered electricity as consumers do not participate sufficiently, and entry barriers in generation might lead to underinvestments. An in-depth game theoretical and extensive empirical analysis would be required to answer this question.4 Neither do I discuss which type of capacity mechanism would be optimal. I show how a simple capacity payment can restore market efficiency, but assume that firms do not behave strategically. An optimal capacity mechanism addresses the market failure it wants to correct, and takes into account the incentives it creates for the participating firms.5 The structure of the chapter is as follows. Sections 8.2 and 8.3 discuss the specificities of electricity markets, explain why a mixture of generation technologies is optimal and how this optimal generation mix can be achieved in a perfectly competitive market using peak-load pricing. Section 8.4 introduces a price cap and shows how it distorts electricity prices, its effect on the profitability of peak and base-load power plants and on long-term investments incentives. Section 8.5 shows that technology-neutral capacity payments can restore investment incentives. Section 8.6 discusses the effect of a larger share of renewable energy and demand side participation on the capacity mechanisms and the generation mix. Section 8.7 concludes.

8.2 Optimal generation mix Any handbook on electricity markets starts from three fundamental characteristics of electricity markets: (a) demand varies over time and is very inelastic, (b) storing other, less onerous forms of regulation that address market failure adequately, such as the introduction of a bid cap, the promotion of bilateral long term contracts, the reduction of entry barriers in generation investments, or the possible divestiture of power plants. Note that Joskow and Tirole show that a price cap does not fully address the market power when there are more than two states of nature. A bid cap might therefore be more appropriate. Paul L. Joskow and Jean Tirole, ‘Reliability and competitive electricity markets’ (2007) The RAND Journal of Economics 38 (1), 60–84. 4 Joskow (n 2) gives an overview of market failures in US markets that could justify the use of capacity markets, but those conditions might not necessarily hold in the EU. For instance, participation in the power exchange is not obligatory (and hence any price cap can be circumvented through bilateral trade), and retail markets are open for competition in the EU (which allows for stronger vertical relationships which reduce market power abuse). Furthermore, it is important that those market failures are explicitly modelled when the design of capacity markets is evaluated. Hogan argues that power markets should not include capacity markets and price caps and that there might be alternative forms of regulation, other than capacity mechanisms, to address market failures, such as insufficient contracts. See William W. Hogan, ‘On an “energy only” electricity market design for resource adequacy’ (Working paper, Center for Business and Government, Harvard University, 2005). Willems and Morbee study the effect of missing financial markets on investment incentives. See Bert Willems and Joris Morbee, ‘Market completeness: How options affect hedging and investments in the electricity sector’ (2010) Energy Economics 32 (4), 786–95; and Bert Willems and Joris Morbee, ‘Risk spillovers and hedging: why do firms invest too much in systemic risk?’ (Katholieke Universiteit Leuven, Centrum voor Economische Studiën, 2011) Center for Economic Studies—Discussion paper. 5 See section 4.3.3 above. For an overview of the economic discussions, see also Leonie Meulman and Nora Méra, ‘Capacity mechanisms in northwest Europe. Between a rock and a hard Place?’ (2012) Clingendael International Energy Programme; and Peter Cramton and Axel Ockenfels, ‘Economics and design of capacity markets for the power sector’ (2012) Zeitschrift Für Energiewirtschaft 36, 113–34.

142

The Generation Mix, Price Caps, and Capacity Payments

electrical energy is expensive, and (c) production capacity is fixed in the short run and capital intensive in the long run. Based on those characteristics, it is then explained how electricity markets rely on a form of peak-load pricing to organize the spot market for electrical energy and how a mixture of base-load and peak-load power plants is optimal.6 I will provide a condensed version of this theory here. Given that electricity cannot be stored, total production needs to be equal to total consumption at each moment in time. This is achieved (a) by constructing a sufficiently large amount of generation capacity such that production output can follow demand most of the time, and (b) by rationing demand when production capacity is insufficient.7 Some production units will only be used for a small number of hours every year when demand is high, while being idle the rest of the year, ie they have a low load factor. For those production units, it is optimal to use production technologies and power plant designs with low capital costs (so-called peak-load power plants or peakers). Other production units will be used for a large number of hours every year and almost never run idle, ie they have a high load factor. For those production units, production technologies and power plant designs that are highly efficient at converting primary energy (gas, coal) into electrical energy, will be employed (so-called base-load power plants). They have higher capital costs than peak-load power plants, but their production efficiency gains outweigh the capital cost disadvantages for high load factors. For a very small fraction of the year even the capital cost of the peak power plants is too high to justify building additional capacity to meet demand. Instead, it is more efficient to ration a small fraction of total demand, and to pay consumers that are rationed to forego electricity consumption entirely. Typically, those compensation payments are relatively high, reflecting the importance of electrical energy, but only have to be paid out during a small number of hours to a small subset of consumers. The price for not receiving electricity is often called VOLL.8 When demand is perfectly inelastic, and neglecting some technical constraints, such as start-up costs, ramp constraints,9 transmission constraints, and the unavailability of generation plants, the optimal generation mix and the optimal amount of rationing can be illustrated graphically. I first sort all 8760 hours of a year, from the hour with the highest level of demand (which is expressed in MW) to the lowest level of demand, and

6

The theory of peak-load pricing and investments dates from Marcel Boiteux. Crew et al and Joskow review the literature on peak-load pricing and describe several extensions of the simplest models. See Marcel Boiteux, ‘The choice of plant and equipment for the production of electric energy’ (1964) in J.R. Nelson (ed), Marginal Cost Pricing in Practice (New York: Englewood Cliffs, Prentice-Hall, 1964) pp 199–2014; Marcel Boiteux, ‘Peak-load pricing’ (1960) The Journal of Business 33 (April), 157–79. Michael A. Crew, Chitru S. Fernando, and Paul R. Kleindorfer, ‘The theory of peak-load pricing: a survey’ (1995) Journal of Regulatory Economics 8 (3), 215–48; Paul L. Joskow, ‘Contributions to the theory of marginal cost pricing’ (1976) The Bell Journal of Economics 7 (1), 197–206. 7 In the text I will use the term ‘demand’ for the amount of energy consumers would like to consume, and ‘consumption’ for the amount of energy consumers actually consume. If demand is rationed, then the consumption will be smaller than the demand. 8 The VOLL is set so that consumers are indifferent between (a) receiving the compensation and (b) receiving electricity. See the definition in chapter 1 at n 7. 9 See the explanation of ramping constraints in chapter 3 at n 4.

8.2 Optimal generation mix 8000

0

1000

2000

3000

4000

5000

143 6000

7000

8000

Demand (MW) Cost (1000s EUR per MW Year)

Demand rationing 6000

300

Peak-load

4000 2000

Base-load

0

250 200 150 100

Base-load Peak-load VOLL

50 0

0

1000

2000

3000 4000 5000 6000 Duration (Hours per Year)

7000

8000

Figure 8.1 Optimal generation mix Source: Author’s own illustration.

represent this in a load duration function.10 One such load duration function is presented in the upper-half of Figure 8.1. The graph shows, for instance, that demand is at least 3,274 MW for 7,000 hours per year, and that the maximal level of demand is equal to 7,961 MW. The total production cost of a generation plant consists of a fixed capital cost, which is incurred even when the power plant is not used, and a variable cost which is proportional to the number of hours the power plant is actually producing electricity. Those variable costs consist mainly of operating and fuel costs. The bottom half of Figure 8.1 presents the total production costs of a base-load plant and a peak-load plant each capable of producing 1 MW, as a function of the number of operating hours in a year, for the parameters shown in Table 8.1. Those lines are also called the screeningcurves. When the power plants do not operate (left side of the figure), they incur only their capital costs, but no operating or fuel costs. Capital costs for the peak-load power plant are €30,000, while they are €100,000 for the base-load power plant. Total production costs will increase with the number of hours that plants operate, but they will increase faster for peak-load than for base-load power plants. At the intersection of those two cost lines (which occurs at 7,000 hours per year) their total costs are equal. The bottom graph shows that a peak-load power plant is cheaper than a base-load power plant whenever it has to run less than 7,000 hours per year. Above this number, the base-load power plant is less expensive. 10 To improve readability of the graph in Figure 8.1, the load duration curve is assumed to have a relatively high downward slope. In practice, load duration functions are often much flatter.

144

The Generation Mix, Price Caps, and Capacity Payments Table 8.1 Parameter values Annualized capital cost €/(MW Year) Peak-load power plant Base-load power plant VOLL Price cap

30,000 100,000

Marginal cost €/MWh 35 25 95 45

Source: Author’s own table.

If demand cannot be rationed, then the total capacity of base-load and peak-load power plants has to be equal to 7,961 MW to meet the highest demand level. So how much of this capacity should be built as base-load power plants, and how much as peak-load power plants? As demand is at least 3,274 MW for 7,000 hours per year, and base-load power plants are cheaper than peak-load power plants if they run more than 7,000 hours per year, we should invest this amount (3,274 MW) in base-load capacity, and allocate the rest to peak-load capacity (4,686 MW = 7,961 MW 3,274 MW). If we can ration demand and compensate consumers for being curtailed, it is efficient to reduce investments in peak-load, and save on investment costs (and some operating costs). Those cost savings will outweigh the disutility of not delivering electricity to a small fraction of consumers for a small fraction of the time. The optimal amount of rationing can be analysed graphically. Demand rationing can be seen as a third ‘production’ technology with zero capital costs and a marginal cost equal to the VOLL. The total cost of rationing 1 MW for a given number of hours per year is indicated by the VOLL line in Figure 8.1. Typical VOLL estimates are in the range 10,000–20,000 €/MWh,11 however, for illustrative purposes, a value of 95 €/MWh is used here.12 This shows that it is inefficient to build peak-load power plants that run for less than 500 hours per year, as it would be more efficient to ration demand instead. So, total installed capacity should only be 6,907 MW, of which 3,274 MW is base-load and 3,633 MW (= 6,907 MW – 3,274 MW) peak-load production. Whenever demand is larger than 6,907 MW, some consumers will be rationed. The optimal generation mix is represented in the second column of Table 8.2. Hence, we have shown that in the social optimum, the generation mix consists of a combination of base-load and peak-load production technologies. The three shaded 11

See explanation of the VOLL in chapter 1 at n 7. The VOLL can directly be observed when energy consumers sign interruptible contracts, or estimated using stated or revealed preference methods, production cost methods, or event studies. In some markets a reference VOLL is set by the regulator. London Economics finds estimates for Britain in the range of 1,600 to 11,820 £/MWh for households, and in the 20,000 to 30,000 £/MWh range for SMEs. De Nooij et al find values around 6,000 €/MWh for Dutch households and 16,400 €/MWh for industry using a production cost method, and Leahy and Tol an average value of 13,000 €/MWh for Ireland and Northern Ireland using the same methodology. See London Economics, The Value of Lost Load (VOLL) for Electricity in Great Britain—Final Report for OFGEM and DECC (July 2013); Michiel de Nooij, Carl Koopmans and Carlijn Bijvoet, ‘The value of supply security: the costs of power interruptions: economic input for damage reduction and investment in networks’ (2007) Energy Economics 29 (2), 277–95; and Eimear Leahy and Richard S.J. Tol, ‘An estimate of the value of lost load for Ireland’ (2011) Energy Policy 39 (3), 1514–20. 12 With a large VOLL, the line would almost become vertical on the graph, and it would be hard to show the intersection of this line with the peak-load generation cost.

8.3 A competitive market leads to an optimal generation mix

145

Table 8.2 Long-term equilibrium investment levels Scenario No price cap Base-load Peak-load Total capacity Maximal demand

3,274 3,633 6,907 7,961

Price cap 3,274 2,233 5,507 7,961

No price cap Wind energy 1,609 4,418 6,027 7,961

Price cap Wind energy 1,609 2,383 3,992 7,961

Source: Author’s own table

areas in the upper half of Figure 8.1 show which combination of the three technologies (base-load, peak-load and demand rationing) is used to ensure that consumption equals supply for each hour of the year. For high demand hours (left side of graph), all three technologies are used to meet demand, for intermediate demand hours (middle of graph), only base-load and peak-load plants are used, while for low demand hours (right of graph), only base-load technology is used.

8.3 A competitive market leads to an optimal generation mix In this section, I will show that the optimal generation mix and its use, as outlined in section 8.2, will be achieved in a competitive electricity market.13 Figure 8.2 shows the market equilibrium for three demand levels. The market price is given by the intersection of the demand and the supply function. Generation firms will produce as long as the spot price is equal to, or above, the marginal production cost. At low demand levels (left graph), the equilibrium price is low (25 €/MWh) and only base-load production is active. At intermediate demand levels (middle graph), the equilibrium price is 35 €/ MWh, and all base-load power plants, and some peak-load plants are producing. At high demand levels (right graph), the price is equal to the VOLL of 95 €/MWh, both peak and base-load produce at full capacity, and a fraction of demand is rationed, as indicated by the arrow. Hence in the market equilibrium, power plants with the lowest marginal cost will always produce before power plants with a higher marginal cost

13

Stoft shows that competitive markets are efficient, as long as prices can rise to the VOLL. The fact that a competitive market equilibrium is efficient is often called the First Welfare Theorem, and has been developed by Arrow, Lange and Lerner. Full efficiency requires that the consumers most willing to pay (with the highest willingness-to-pay) receive the energy and the consumers least willing to pay (with the lowest willingness-to-pay) are rationed first. Technically, this is only feasible if smart meters are installed, and the distribution operator can curtail consumption at the household level. With older technologies, curtailment is only possible at regional or district level, and full efficiency is not possible. However, Joskow and Tirole show that a competitive market equilibrium is constrained efficient, ie given the imperfect rationing technology, the market outcome cannot be improved upon. See Steven Stoft, Power System Economics: Designing Markets for Electricity (New York: IEEE Press and Wiley-Interscience, 2002); Kenneth J. Arrow, ‘An extension of the basic theorems of classical welfare economics’ (The Regents of the University of California, 1951); Oscar Lange, ‘The foundations of welfare economics’ (1942) Econometrica 10 (3), 215–28; Abba P. Lerner, ‘Economic theory and socialist economy’ (1934) The Review of Economic Studies 2 (1), 51–61; and Joskow and Tirole (n 3).

The Generation Mix, Price Caps, and Capacity Payments 100

100

90

90

90

80

80

80

70 60 50 40

Price (Eur/MWh)

100

Price (Eur/MWh)

Price (Eur/MWh)

146

70 60 50 40

70 60 50 40

30

30

30

20

20

20

10

10

10

0 0

2000 4000 6000 8000 Quantity (MW)

0 0

2000 4000 6000 8000 Quantity (MW)

Market Equilibrium

Demand

0 0

2000 4000 6000 8000 Quantity (MW)

Supply

Figure 8.2 Market equilibrium for low, medium, and high demand Source: Author’s own illustration.

come online. This is efficient as it minimized costs, and corresponds to the three shaded areas laid out in Figure 8.1. Firms will invest as long as they can make some positive or zero profit.14 Hence, at long-term equilibrium investment levels, firms will make zero profit on base-load power plants and zero profit on peak-load power plants. We will now show that the investment quantities that correspond to this market equilibrium are the ones derived in section 8.3. In other words, given the investment levels in column 2 of Table 8.2, generators make zero long-term profit on peak-load and base-load generation capacity. As it might not be immediately obvious, I will now discuss why this is the case. There are 500 hours a year in which demand is rationed, and during those hours, the electricity price is equal to 95 €/MWh, as shown in the right graph of Figure 8.2. This price is exactly equal to the compensation that needs to be paid to consumers for rationing demand. However, for a load factor of 500 hours per year, the total production cost of peak-load production is equal to the total cost for rationing consumers, as can be seen by the intersection of the total cost lines in the bottom half of Figure 8.1. Hence, the price 95 €/MWh is also sufficient to pay for the total cost (investment cost plus variable cost) of peak producers operating 500 hours per year. If peak producers operate more than 500 hours per year, then they receive 95 €/MWh during the 500 hours with the highest demand and a price equal to their own marginal cost (35 €/ MWh) during the remainder of the hours. Hence, the first 500 hours are sufficient to pay for investment and for the variable costs, while the additional revenue for the hours above 500 is equal to the additional production cost. Peak producers therefore make zero profit. 14

Note that profit is defined differently by accountants and economists. Accountants calculate profit without taking into account a cost for the capital provided by shareholders. Economists calculate profit taking account an appropriate compensation of this capital. Here, we look at the economists’ concept.

8.4 A price cap distorts investment levels

147

If base-load generators operate 7,000 hours per year, their total cost is equal to the total cost of peak-load generators. The revenue that a peaker earns during those 7,000 hours is equal to the peaker’s total costs (as it has zero profit); hence, it is also exactly equal to the total cost of the base-load generator. If the base-load generator produces more than 7,000 hours per year it will receive a price for additional production which is equal to its marginal production costs. Therefore, the base-load generator will also have zero profit on its investment quantities in the long run (given the scenario with no price cap, shown in the second column of Table 8.2). Summarizing, in a competitive electricity market, private investment and production decisions are such that the optimal generation mix will be achieved.

8.4 A price cap distorts investment levels This section studies the short-term and long-term equilibrium effects of a price cap on a competitive electricity market.15 I assume a price cap of 45 €/MWh, as indicated in Table 8.1. The introduction of the price cap will lower electricity prices during peak hours from 95 €/MWh to 45 €/MWh. This price reduction will benefit consumers at the expense of producers. The price cap will reduce revenue for all generation plants by 50 €/MWh for 500 hours per year. This creates a yearly shortfall of 25,000 € per MW installed capacity, both for peak-load and base-load generation plants. This shortfall, discussed at length in chapters 1, 4, and 5 above, is called the missing money problem. In the short run, investment levels remain constant and are not affected by the price cap. Short-term supply is perfectly inelastic at the price cap, and demand is also assumed to be perfectly inelastic.16 Hence, there are no short-term negative welfare effects of introducing a price cap. Myopic policy makers (or network operators) might be tempted to impose such a price cap to transfer revenue from generators to consumers without any obvious negative effects on market outcomes.17 In the long run, supply is perfectly elastic, as we assume free market entry, and the price cap will therefore distort the market equilibrium and create deadweight losses. Given the missing money problem, generators are unable to cover their capital costs, and installed capacities will decrease until the resulting higher prices are sufficient to pay for the capital costs. In the long run, the market will reach a new equilibrium, in which all generators break-even. As those capital costs are sunk, power plants might 15 Here, I study the effect of a price cap on investments in a competitive situation. Buehler et al study the effect of a price cap on investments in a Cournot oligopoly with one technology and fluctuating demand. Zöttl introduces a base-load and a peak-load technology and shows that a well-chosen price cap might increase investments in an oligopoly model as the incentives for withholding capacity are reduced. Stefan Buehler, Anton Burger, and Robert Ferstl, ‘The investment effects of price caps under imperfect competition: a note’ (2010) Economics Letters 106 (2), 92–4. Gregor Zöttl, ‘On optimal scarcity prices’ (2011) International Journal of Industrial Organization 29 (5), 589–605. 16 In our model, short-term supply is perfectly inelastic as long as the price cap is higher than the marginal cost of the peak-load power plant. 17 In practice, production capacity and demand are not perfectly inelastic, and a price cap might therefore create some deadweight losses. For instance, a price cap might reduce incentives for generators to schedule maintenance in periods where demand is likely to be low.

148

The Generation Mix, Price Caps, and Capacity Payments

not immediately be mothballed after the introduction of a price cap, as they are still making a profit on the margin, but power plants might retire earlier, and not all retired power plants will be replaced. We can determine the long-term equilibrium capacities graphically by lowering the value of VOLL to the level of the price cap, as illustrated in Figure 8.3, and use the same analysis as in section 8.2. Figure 8.3 shows that peak-load capacity is reduced to the point where consumers are rationed about 3,000 hours per year. At this level, prices are sufficiently high to pay for the total cost of the peak-load capacity. The lower price during hours with demand rationing (45 €/MWh instead of 95 €/MWh) is offset by an increase of the duration of demand rationing hours (from 500 to 3,000 hours per year). The markup of 10 €/MWh, which the peak-load power plant earns on top of its marginal cost during those 3,000 hours, is sufficient to pay for its yearly capital cost (10 €/MWh times 3,000 hours per year, or 30,000 €/year). The total amount of base-load capacity remains, maybe somewhat surprisingly, unchanged. As the peak power plants break-even at a duration of 7,000 hours per year and as the total production cost for base-load is identical to peak-load at this load factor, the base-load power plants will also break even. Hence, although in the short run, both the peak and the base-load power plants face the missing money problem, in the long run, investment levels will decrease only in peak-load capacity. The long-term equilibrium investment levels are given in column 3 of Table 8.2.

Demand (MW)

6000

Cost (1000s EUR per MW Year)

8000

300

0

1000

2000

3000

4000

5000

6000

7000

8000

4000 2000 0

250 200 150 100

Base-load Peak-load Price Cap

50 0

0

1000

2000

3000 4000 5000 6000 Duration (Hours per Year)

Figure 8.3 Long-term effect of a price cap Source: Author’s own illustration.

7000

8000

8.5 Capacity payments restore efficiency

149

As long-term supply is elastic, the price cap distorts investment decisions, and creates a deadweight loss. Hence, from a social viewpoint total welfare is reduced, and demand is rationed too often.

8.5 Capacity payments restore efficiency The previous section has shown that a price cap will lower peak-load capacity, increase demand rationing, and lower overall welfare. I will now discuss how a capacity payment can restore market efficiency. The introduction of the price cap lowered the VOLL-line in Figure 8.1 to the price cap line in Figure 8.3. As a result, the intersection of this line with the total cost line for peak-load capacity moved to the right, and the number of hours with rationing increased from 500 to 3,000 hours per year. By subsidizing investments in peak-load capacity using a capacity payment, the total cost line for peak-load capacity will shift downwards, and the intersection with the price cap line will move to the left, as illustrated by Figure 8.4 (bottom half ). If we give a capacity payment of €25,000 per year to peak-load generators, then the intersection with the price cap line will be at 500 hours per year and the number of hours with rationing reaches the optimal level again. However, a subsidy to the peak-load generators will also affect the trade-off between base-load and peak-load generation. Peakload generation becomes relatively cheaper than base-load generation, and would be operating more than 7,000 hours per year as the intersection between the total cost line of peak-load and base-load would shift to the right. Therefore, in order to obtain the

Demand (MW)

6000

Cost (1000s EUR per MW Year)

8000

300

0

1000

2000

3000

4000

5000

6000

7000

8000

4000 2000 0

250 200 150 100

Base-load w/ capacity payment Peak-load w/ capacity payment Price Cap

50 0

0

1000

2000

3000 4000 5000 6000 Duration (Hours per Year)

Figure 8.4 Price cap with capacity payment Source: Author’s own illustration.

7000

8000

150

The Generation Mix, Price Caps, and Capacity Payments

optimal mixture of peak- and base-load generation, base-load generation should also receive the same capacity payment of €25,000 per year. This will shift the total cost line of base-load capacity downwards with the same amount as the peak-load plant, such that it intersects at the optimal level of 7,000 hours per year (see Figure 8.4). Hence, we have shown that a technology-neutral capacity payment, which is equal to the amount of missing money (25,000 €/MWh per year), will correct investment incentives. If the capacity payment were only given to peak-load power plants, then too few base-load power plants would be built.

8.6 Renewable energy and demand participation This section discusses the effect of renewable energy and demand participation on the generation mix. I start by assuming that some producers with wind farms are present in the system. As wind has a marginal cost equal to zero and will always be used whenever it is available, we can subtract it from the total demand level, and determine the optimal generation mix of the remaining technologies on net demand, that is demand minus wind output. Figure 8.5 shows the load duration curve of net demand after the introduction of wind energy, under the assumption that that wind production is uniformly distributed between 0 and 3,000 MW and is not correlated with demand. The maximal level of demand is still as high as before, as there is some probability that there will be no wind in the hour with the highest demand. There are also a number of hours where

Demand (MW)

8000

0

1000

2000

3000

4000

5000

6000

7000

8000

6000 4000 2000

Cost (1000s EUR per MW Year)

0 300 250 200 150 100

Base-load Baseload Peakload Peak-load VOLL

50 0

0

1000

2000

3000 4000 5000 6000 Duration (Hours per Year)

Figure 8.5 Optimal generation mix with 3,000 MW wind energy Source: Author’s own illustration.

7000

8000

8.6 Renewable energy and demand participation

151

production by the wind farm is sufficient to satisfy total demand, and other generation is not needed to satisfy demand.18 The optimal generation mix can be derived as before. Demand will be rationed 500 hours per year, and base-load will only operate if it can operate more than 7,000 hours per year. The investment levels will, however, be adjusted in accordance with the new net load duration function. Base-load capacity reduces from 3,274 to 1,609 MW, while peak-load capacity increases from 3,633 to 4,418 MW. Total capacity reduces from 6,907 to 6,027 MW (compare column 2 and column 4 in Table 8.2). In the short run, the addition of wind energy production reduces the revenue of both base-load and peak-load generation plants. Those power plants will see lower load factors, and will no longer be able to recover their investment costs. In the long run, base-load plants will be mothballed, and only a fraction of them will be replaced by new peak-load power plants. As a result, total capacity will decrease. Note that, although firms are unable to recover their capital cost in the short run, there is no need to introduce a capacity mechanism, as those low profit levels are an indication that total capacity needs to be reduced and that the market is undergoing a structural reform towards a low carbon economy.19 As capital costs are sunk, there will not be an immediate mothballing of existing capacity; the adjustments are expected to be gradual. As base-load capacity has a longer life time and a lower marginal cost than peak-load capacity, some unexpected transition effects might arise. Peak-load capacity is mothballed first, but not replaced, because there is still overcapacity in the market (in the form of base-load capacity). So, in the intermediate term, the fraction of base-load capacity in the generation mix might actually increase. Only at a later stage, when baseload capacity is no longer economically viable and is being phased out, one might see a gradual shift of the generation mix towards peak-load power plants. Such a transition stage, where new peak-load capacity is only built when sufficient base-load capacity has retired, is likely to be socially optimal, but might create some unease among policy makers, as there is little new peak power capacity being built in the transition phase.20 We can analyse the effect of a price cap in the model with wind energy in the same manner as in the model without it (see section 8.2), and would see that the price cap will affect investment levels to a larger extent, as the load duration curve is now steeper. The price cap reduces peak capacity by 2,035 MW in the situation with wind energy (column 4 minus column 5 in Table 8.2) and by 1,400 MW in the situation without it (column 2 minus column 3 in Table 8.2). However, the same technology-neutral capacity payment of €25,000 per year is required to correct incentives, as the amount of missing money has not changed (500 hours per year times 50 €/MWh).

18 Kennedy (2005) studies the effects of wind energy on the long-term investment strategies using a similar method as the one used here, but takes into account the correlation between wind production and demand in calculating the net load duration function. See Scott Kennedy, ‘Wind power planning: assessing long-term costs and benefits’ (2005) Energy Policy 33 (13), 1661–75. 19 In this situation, firms typically will have a lower accounting profit and lower dividends, and a negative economic profit, as defined at n 14. 20 However, this fear might be justified if there are other market distortions. For instance, peak power plants might receive a price too low in the balancing market for providing flexibility.

152

The Generation Mix, Price Caps, and Capacity Payments 0

1000

2000

3000

4000

5000

6000

7000

8000

6000 4000 2000

Cost (1000s EUR per MW Year)

Demand (MW)

8000

0 300 250 200 150 100

Base-load Peak-load VOLL

50 0

0

1000

2000

3000 4000 5000 6000 Duration (Hours per Year)

7000

8000

Figure 8.6 Effect of demand elasticity Source: Author’s own illustration.

In this chapter, I have assumed that demand is perfectly inelastic. This allows us to derive the equilibrium generation mix graphically. When demand is elastic (for instance, due to the introduction of a smart meter), the determination of the optimal generation mix can no longer be represented in a simple graph. The load duration function now depends on the price level and therefore on the generation mix.21 The load duration curve will be higher if prices are low (when base-load is producing) and lower if prices are higher (when peak-load is producing). The resulting load duration function is a combination of the high and low price load duration functions, as shown in the upper half of Figure 8.6. The load duration function will become flatter and the relative fraction of base-load in the generation mix will increase at the expense of peak-load capacity. Hence, demand participation might offset the effect of renewable energy, which made the load duration function steeper. The effect of a price cap will have a similar effect on the generation mix as when demand is inelastic. In the long run it will lower investments in peak-load capacity and in order to restore efficiency, a technology-neutral capacity payment needs to be made both to base- and peak-load capacity.

21 Efficient investment levels cannot be determined by the graphical approach and need to be evaluated numerically. However, the market outcome can still be represented graphically. Note that the graph assumes that demand is elastic for intermediate price levels, but inelastic for high price levels. For the effect of risk aversion in input cost uncertainty using a graphical analysis, see Guy Meunier, ‘Risk aversion and technology mix in an electricity market’ (2013) 40 Energy Economics 40 (C), 866–74.

8.7 Conclusion

153

The only difference, with respect to the previous analysis, is that if demand is priceelastic, the price cap will also create a deadweight loss in the short run, as consumers will receive the wrong incentives to restrict demand.

8.7 Conclusion This chapter provides an economic analysis of a capacity payment mechanism in a simplified perfectly competitive electricity market, in order to highlight some key concepts of electricity market design. Given the inability to economically store electricity, energy consumption and energy production should be balanced at each moment in time. Such a balance can be achieved by producing electricity with peak-load and base-load power plants, and by rationing demand. A perfectly competitive market ensures both short- and long-term efficiency. In the short run, available technologies are used in an efficient way to balance demand and supply. Technologies will be used according to their marginal costs, ie according to the merit order.22 Base-load technologies are used first, followed by peak-load technologies, and as a last resort, demand rationing. For long-term equilibrium, firms will invest in an optimal combination of peak- and base-load generation technologies. Myopic policy makers might be tempted to introduce a price cap in the spot market, as it reduces electricity prices for consumers without causing any welfare losses in the short term. However, it will reduce the revenues for generators, who will no longer be able to pay for their investment costs; a phenomenon which is called the missing money problem. In the long run, a price cap will result in an underinvestment in peakload power plants and lower welfare levels, as there is too much demand rationing. In order to restore efficiency, a technology-neutral capacity payment can be introduced. The shift towards a low carbon electricity system, for instance by the further integration of wind energy, will require a reduction of the total installed capacities of other technologies and a generation mix which relies more on peak-load power plants than on base-load power plants. In the transition phase, some power plants will no longer be able to cover their capital costs. However, this is not a reason to introduce a capacity payment (there is no missing money). Rather, the fact that some power plants cannot recover their costs is a consequence of a structural reform of the generation system which makes some power plants obsolete. As base-load power plants have a lower marginal cost and a longer economic lifetime than peak-load power plants, we might observe some unexpected dynamics in the transition: an initial reduction of peak-load power plants, followed (only at a later stage) by a gradual replacement of base-load power plants by peak-load power plants. Hence, the larger share of peakload generation in the system might only be achieved after this transitional phase has passed. Introducing a price cap in a low carbon electricity system will affect investment levels to a larger extent than introducing the same price cap in a system without renewable energy. However, efficiency can again be restored with a technology-neutral 22

See sections 1.1, 2.2, and 4.2.1 for a discussion.

154

The Generation Mix, Price Caps, and Capacity Payments

capacity payment, and the rate of this payment is independent of the quantity of renewable energy in the system. Demand participation will offset the effects of wind integration, make the resulting load duration curve flatter and lead to a generation mix which relies more on base-load and less on peak-load capacity. Note that this chapter does not discuss (a) whether a capacity mechanism is necessary for a well-functioning power market, and (b) which type of capacity mechanism is best suited to address a particular market failure. With respect to the first question, there is no consensus yet on whether market failures in the electricity market justify government intervention, as there are many other markets with capital intensive industries which seem to function efficiently without capacity mechanisms. With respect to the second question, some market-based capacity mechanisms such as reliability options might provide better incentives than administratively imposed capacity payments. These different capacity mechanisms are discussed in greater detail in Chapter 1, section 2.3 above.

PART III LAW

9 Capacity Mechanisms and State Aid Control A European Solution to the ‘Missing Money’ Problem? Leigh Hancher

9.1 Introduction Section 9.2 of this chapter examines the evolution of the Commission’s policy on the application of the European state aid rules to various forms of capacity mechanisms. How do these rules apply to national measures adopted to confront the missing money problem discussed extensively in the previous chapters of this book?1 As will become evident, the application of the EU state aid rules—Articles 107 and 108 TFEU—has been recognized in various official documents released by the Commission as being of key importance to ensuring that the introduction of different forms of financial support to ensure that generation adequacy do not frustrate the over-riding objective of completing the internal energy market by December 2014. On 22 May 2013 the European Council called for particular priority to be given to the Commission providing guidance on capacity mechanisms.2 The Commission subsequently met this call with the November 2013 Communication3 and the adoption of the EEAG 2014–20204 which includes specific criteria for assessing the compatibility of state aid to support generation adequacy measures.5 Important as these binding Treaty rules may be, one should not over-estimate their potential impact. It is always possible for Member States to design support schemes which do not fall within the scope of the EU state aid rules at all, especially if those support schemes are not financed by or through state resources. The classification of a support scheme or measure as state aid is addressed in section 9.3 of this chapter. Even if the measures in question are to be designated as state aid within the meaning of Article 107(1) TFEU, the Commission’s primary role is to consider whether the state aid measure as notified is compatible aid in accordance with Article 107(3)(c) TFEU. The Commission does not design the national measure but assesses it in accordance with certain criteria, now laid down in ‘soft-law’ guidelines in the form of the EEAG 2014–2020. 1

See in particular chapters 1, 2, and 5. European Council, Conclusions (EUCO 75/1/13 REV 1, 22 May 2013) in particular point 2, available at: http://www.consilium.europa.eu/ueDocs/cms_Data/docs/pressData/en/ec/137197.pdf, accessed 1 February 2015. 3 Communication from the Commission, Delivering the internal electricity market and making the most of public intervention, C (2013) 7243 final (November 2013 Communication). 4 Communication from the Commission, Guidelines on State aid for environmental protection and energy 2014–2020 [2014] OJ C 200/1–55 (EEAG 2014–2020). 5 On 25 April 2015 the Commission launched its first sector inquiry into capacity mechanisms (IP-154891). 2

158

Capacity Mechanisms and State Aid Control

Obviously then, the application of the EU state aid rules are not an alternative to or substitute for a harmonized approach to dealing with generation adequacy at European level. The EU’s role is reactive and even passive: at most it can reject a national capacity mechanism scheme as incompatible state aid, but on the basis of past precedent, it is more likely that the Commission will recommend certain changes or adjustments in order to approve the scheme in question as compatible aid. Furthermore the Commission considers each national measure individually and on its own merits. It does not compare that measure with the measures adopted in another Member State in order to determine compatibility—state aid control is not even an instrument of indirect harmonization. The launch of the Sector Inquiry in April 2015 will not alter this state of affairs. It is therefore important to consider how far-reaching the criteria developed by the Commission in its soft-law guidelines are likely to be in practice in realizing the overall objective of securing the functioning of the internal energy market. This topic is considered in section 9.4. In section 9.5, the chapter offers an assessment of the application of the EEAG 2014–2020 so far, and then draws a number of conclusions in section 9.6.

9.2 Policy evolution Public intervention in the energy sector of the twenty eight Member States is by no means novel. As the results of a study released by the Commission in October 2014 show, in 2012, the total value of public intervention in energy (excluding transport) in the EU twenty eight was between €120 and €140 billion. Unsurprisingly, and given the efforts to expand the share of renewable energy in the EU’s overall energy consumption, the largest amounts of current public support in 2012 went to renewables, in particular to solar (€14.7 billion) and onshore wind (€10.1 billion), followed by biomass (€8.3 billion) and hydropower (€5.2 billion). Among conventional power generation technologies, coal received the largest amount in current subsidies in 2012 with €10.1 billion, followed by nuclear (€7 billion) and natural gas (about €5.2 billion).6 This recent study does not however include the cost of capacity mechanisms.7 Capacity mechanisms come in various guises, as discussed extensively elsewhere in this book.8 They also come at a cost. International experience shows that capacity mechanisms can cost up to 10–20 per cent of wholesale electricity (ie energy only) prices.9 The UK government has estimated that the costs of its capacity mechanism, discussed in detail in chapter 22, will be in excess of £2 billion per year. If that cost is borne by the state or financed through state resources, then the EU state aid rules are likely to come into play. 6 The figures specifying support across technologies do not reflect the free allocation of emission certificates nor tax support for energy consumption. 7 Ecofys, Subsidies and costs of EU energy—final report, study prepared for the Commission (project DESNL14583, 1 November 2014). See also European Commission, Press release IP/14/1131, 13 October 2014. 8 For an overview of different capacity mechanism types, see section 1.2.3 above. 9 Commission staff working document, Generation Adequacy in the internal electricity market— guidance on public interventions, 5 November 2013 (Generation Adequacy SWD) p 32.

9.2 Policy evolution

159

The adoption of generation support mechanisms is not necessarily a recent phenomenon, although it has become a major European energy policy issue in the last two years. There are examples of national mechanisms to support investment being introduced following the adoption of the 2003 Electricity Directive,10 as discussed in greater detail in chapter 1. Prior to the gradual introduction of wholesale electricity market liberalization Member States were allowed to compensate their electricity generators for ‘stranded costs’, that is for the so-called ‘non-returned’ investment costs incurred in constructing generating plant prior to market liberalization that could no longer be passed on to final customers in a competitive market. The Commission recognized at the time that energy market transitions—in this case from a regulated to a de-regulated ‘energy only’ market—came at a substantial cost. That cost could effectively be socialized so that compensation for ‘non-returned investments’, albeit framed in accordance with the Commission’s so-called Stranded Costs Communication of 2001, could be justified under the EU state aid rules.11 If the EU’s electricity market legislation aimed at achieving a transition from a regulated to a de-regulated generation market, its legislation on RES seemed to travel in the opposite direction. Energy market transition—this time to a low-carbon based system—has again come at a cost. With the adoption of the 20/20/20 Package12 and the introduction of binding national targets for RES, capacity mechanisms may be deemed necessary to secure new investment in power plants as well as generation flexibility.13 This latter rationale has become more prominent: capacity mechanisms may be required to incentivize thermal capacity to remain on the national electricity system and provide back-up supplies when intermittent RES production fails to deliver. The increased share of RES in the low-carbon energy market creates a new type of market failure as flexible backup capacity has little financial incentive to remain in the market.14 Back in 2001 the Commission attempted to apply the EU state aid rules to secure market entry for new investors. Its complex methodology set out in the Stranded Costs Communication was designed to allow compensation to be paid to incumbent generators whose plant would not be competitive in a liberalized market, but only up to a level that would not prevent new investment or if adequate interconnection was in place, imports of electricity. The vast majority of the Member States provided some 10 Directive 2003/54/EC of the European Parliament and of the Council of 26 June 2003 concerning common rules for the internal market in electricity and repealing Directive 96/92/EC [2003] OJ L 176/37 (2003 Electricity Directive). 11 Commission Communication relating to the methodology for analysing State aid linked to stranded costs, adopted by the Commission on 26 July 2001, Commission letter SG (2001) D/290869 of 6 August 2001 (Stranded Costs Communication). 12 The climate and energy package (the 20/20/20 Package) is a set of binding legislation which aims to ensure the EU meets its ambitious climate and energy targets for 2020. These targets, known as the ‘20/20/ 20 targets’, set three key objectives for 2020: (a) a 20% reduction in EU greenhouse gas emissions from 1990 levels; (b) raising the share of EU energy consumption produced from RES to 20%; (c) a 20% improvement in the EU’s energy efficiency. For more information on the Package, go to http://ec.europa.eu/clima/ policies/package/index_en.htm, accessed 1 February 2015. See above, section 1.4. 13 See Communication from the Commission, Renewable Energy: a major player in the European energy market, COM (2012) 271 final. 14 See above, sections 1.1 and 4.4.

160

Capacity Mechanisms and State Aid Control

form of stranded cost compensation in accordance with the Stranded Costs Communication. Energy market transition, it would seem, usually comes at a heavy price ultimately borne by end consumers. The latest transition challenge to which the EU state aid rules must rise to is not so much to smooth the transition to a (low carbon) Europe-wide energy-only market but to prevent the collapse of that very objective. This is where the ‘missing money’ problems come in: liberalized wholesale markets may not remunerate the invested capital at the rates required by the market to invest. Yet capacity mechanisms are not part of the Commission’s Target Model for electricity, with its emphasis on cross-border trade and day-ahead and intraday markets.15 Rather, they are more readily perceived as a response to its shortcomings and inconsistencies, and in particular the failure to complete the internal market for renewable energy.16 In accordance with the Security of Supply Directive17 public authorities must regularly undertake an objective, facts based, assessment of the generation adequacy situation in their Member State fully taking account of developments at regional and Union level. The Commission has also acknowledged that the rules contained in the Security of Supply Directive and its transposition and implementation may be insufficient to tackle the challenges of the future in a fully satisfactory way. Therefore, as set out in the November 2013 Communication, the Commission may consider taking further legislative initiatives in this regard—although to date this has not occurred.18 In the absence of any other suitable weapons the Commission is left with the choice of pursuing infringement actions in the event that capacity mechanisms breach the EU rules on free movement and/or mobilizing the EU state aid rules. On the face of it the latter strategy has its attractions. Member States must notify their plans to grant aid and are bound by the ‘standstill’ provision in Article 108(2) TFEU to await a positive assessment of those plans by the Commission before putting them into effect. Potential beneficiaries prefer the security of a positive state aid decision as opposed to the threat of a recovery order. Recalcitrant Member States can eventually be sanctioned on the basis of fines. In the majority of cases however, the Commission and the Member State adjust the notified aid measures so that the Commission can adopt a positive decision, allowing the aid, but subject to conditions which better reflect its more general European policy aims. In parallel with its case-by-case assessment of national capacity mechanisms under the EU state aid rules, the Commission is undertaking detailed studies on the development of a European generation and system adequacy assessment. These will help identify the adequacy standards which are appropriate in an effectively functioning internal energy market. This work will involve ENTSO-E,19 ACER, and the Member 15

16 See above, section 5.3.1. See chapter 11. Directive 2005/89/EC of the European Parliament and of the Council of 18 January 2006 concerning measures to safeguard security of electricity supply and infrastructure investment [2006] OJ L 33/22 (Security of Supply Directive) Art 4, specifically requires Member States not to discriminate between cross border contracts and national contracts. 18 November 2013 Communication (n 3) p 13. 19 The European Network of Transmission System Operators for Electricity (ENTSO-E), an association of Europe’s transmission system operators (TSOs) for electricity, produces Union-wide generation adequacy assessments in accordance with Art 8 of Regulation (EC) 714/2009 of the European Parliament 17

9.2 Policy evolution

161

States’ authorities, including through the Electricity Coordination Group. It is intended that the outcome of these studies will provide an objective evidence base for future work by the Commission.20 In the absence of further harmonizing measures or common standards on adequacy assessment, however, state aid control applied on a case-by-case basis will be a primary if not the sole instrument to achieve the goals envisaged in the 2013 Communication. Even if harmonizing measures were to be adopted, this would of course not exclude the application of the Treaty rules on state aid. It is therefore important to focus on how effective this instrument can prove in practice.

9.2.1 State Aid Modernisation (SAM) The adoption of the new EEAG 2014–2020 to replace the earlier Environmental Aid Guidelines 2008 (EAG 2008) which expired at the end of 201321 has allowed the Commission to incorporate specific compatibility criteria for the assessment of capacity mechanisms under the state aid rules. It should however be pointed out that the overall design of the EEAG 2014–2020 is also a result of the State Aid Modernisation exercise (SAM), launched in May 2012.22 The then Commissioner Almunia declared that reforms should meet three main, closely linked objectives: (a) foster growth in a strengthened, dynamic, and competitive internal market, (b) focus enforcement on cases with the biggest impact on the internal market, and (c) provide streamlined rules and faster decisions. The Commission considered that state aid control should more effectively target sustainable growth-enhancing policies while encouraging budgetary consolidation, limiting distortions of competition and keeping the single market open. Further, in order to enhance the coherence of state aid control the Commission resolved to identify common principles for assessing the compatibility of aid with the internal market, across various guidelines and frameworks and to revise, streamline, and consolidate state aid guidelines to make them consistent with those common principles. Importantly, the Commission perceives state aid control as an instrument for national budgetary discipline: Modernised State aid control should facilitate the treatment of aid which is welldesigned, targeted at identified market failures and objectives of common interest, and least distortive (‘good aid’). This shall ensure that public support stimulates and of the Council of 13 July 2009 on conditions for access to the network for cross-border exchanges in electricity and repealing Regulation (EC) 1228/2003 [2009] OJ L 211/15 (2009 Cross-border Regulation). ENTSO-E’s Report is currently the main Union-wide assessment of generation adequacy. See the most recent report of ENTSO-E, Scenario Outlook and Adequacy Forecast 2014–2030, June 2014. See also the further work of the Electricity Co-ordination Group and ENSO-E, available at http://www.ec.europa.eu/ competition/sectors/energy/state_aid_to_secure_electricity_supply_en.html, accessed 7 July 2015. 20 Communication from the Commission, Progress towards completing the Internal Energy Market, COM (2014) 634 final, 13 October 2014 (October 2014 Communication). 21 Notice from the Commission, Community guidelines on state aid for environmental protection [2008] C 82/1. 22 Communication from the Commission, EU State aid modernisation (SAM), COM 2012 (209) final, 8 May 2012.

162

Capacity Mechanisms and State Aid Control

innovation, green technologies, human capital development, avoids environmental harm and ultimately promotes growth, employment and EU competitiveness.23

The aims of the modernization exercise are reflected in the new EEAG, which now includes specific guidelines on certain forms of aid to the energy sector. Before turning to the content of the guidelines it is important to determine the scope of the EU state aid regime in order to establish when national measures could be subject to them.

9.3 Funding capacity mechanisms: When do the state aid rules apply? Although the new EEAG, discussed in detail in section 9.4 below, provide us with guidance on when state support for capacity mechanisms may or may not be compatible with the internal market, the important question of whether the EU state aid rules are applicable to such support measures in the first place is not addressed there. This is broadly typical of the Commission’s approach as its guidelines focus primarily on compatibility. One may note here that certain sets of Commission guidelines do in fact provide some guidance as to the application of the market investor principle—for example in the revised Broadband Guidelines of December 2012,24 the revised Risk Capital Guidelines of 2014,25 and the revised Airports and Airlines Guidelines.26 The EEAG 2014–2020 does not follow this pattern however. Further, the final version of the EEAG, and again in contrast to other sets of recent guidelines, is silent on the application of Article 106(2) TFEU as a possible ground for compatibility of the measure, in relation to SGEI.27 In an early non-published draft version of EEAG of July 2013, the Commission had considered that ‘in general terms, [ . . . ] Member States cannot attach specific public service obligations to services that are provided or can be provided satisfactorily and under conditions, such as price, objective quality characteristics, continuity and access to the service, consistent with the public interest, as defined by the State, by undertakings operating under normal market conditions.’28 Therefore the Commission concluded that ‘the provision of services such as a specific level of electricity generation adequacy is unlikely to be a compatible form of support for the performance of an SGEI within the meaning of Article 106(2)

23

SAM (n 22) para 12. Communication from the Commission, EU Guidelines for the application of State aid rules in relation to the rapid deployment of broadband networks [2013] OJ C 25/1. See its sections 2.1 and 2.2. 25 Communication from the Commission, Guidelines on State aid to promote risk finance investments [2014] OJ C 19/4. See its section 2.1. 26 Communication from the Commission, Guidelines on State aid to airports and airlines [2014] OJ C 99/3. See its section 3.1. 27 Services of General Economic Interest (SGEI) are commercial services of general economic utility, for example transport, energy and communications services, on which public authorities impose public service obligations (PSOs): http://europa.eu/legislation_summaries/glossary/services_general_economic_interest_ en.htm, accessed 1 February 2015. 28 Paper of the services of DG Competition containing draft Guidelines on environmental and energy aid for 2014–2020 (draft EEAG 2014–2020, non-published version of July 2013) para 28. 24

9.3 Funding capacity mechanisms: When do state aid rules apply?

163

TFEU. The provision of electricity is under normal circumstances a service provided by undertakings competing in a market, and such service cannot be easily distinguished from the provision of capacity.’29 The later draft version of EEAG issued for consultation on 18 December 201330 already confirmed that the Commission had changed its mind. It recognized that Member States may consider that certain services provided in the energy sector should be regarded as SGEI within the meaning of Article 106(2) TFEU, and went on to indicate that any compensation for the provision of SGEI would be assessed in line with the Commission’s general ‘SGEI Package’ as adopted on 20 December 2011.31 Given that state intervention in the energy sector has been a constant fact of European life since the creation of the first European Coal and Steel Community (ECSC) followed by the Euratom Community, there is a wealth of jurisprudence on which to draw in determining whether a particular form of state intervention in the energy sector may fall within the scope of Article 107(1) TFEU. Yet this jurisprudence is by no means clear and has not yet fully resolved the controversies of classification. Prior to the adoption of the EEAG 2014–2020 there were very few rulings on capacity mechanisms,32 but the recent case law on feed-in tariffs and other forms of support for renewable energy sources can provide a valuable guide for Member States attempting to design capacity mechanisms that fall outside the reach of the EU state aid rules.

9.3.1 Article 107(1) TFEU As is well known, in order for a measure to be classified as a state aid within the meaning of Article 107(1) TFEU it must fulfil four cumulative conditions. There must be (a) an economic advantage, favouring certain goods or sectors, (b) funded by or through state resources which (c) distorts competition and (d) adversely affects trade within the internal market. The European Courts have consistently held that Article 29 Draft EEAG 2014–2020, non-published version of July 2013 (n 28) para 29. Similarly the Commission dismissed the possibility that the activities of TSOs and DSOs (distribution system operators) could be designated as SGEIs (at para 31). 30 Paper of the Services of DG Competition containing draft Guidelines on environmental and energy aid for 2014–2020 (18 December 2013), available at http://ec.europa.eu/competition/consultations/2013_ state_aid_environment/draft_guidelines_en.pdf, accessed 1 February 2015 (draft EEAG 2014–2020, version of 18 December 2013). 31 The SGEI package consists of four documents. The first three documents were adopted on 20 December 2011 and include: (1) Communication from the Commission on the application of the European Union State aid rules to compensation granted for the provision of services of general economic interest [2012] OJ C 8/4–14, (2) Commission decision 2012/21/EU of 20 December 2011 on the application of Article 106(2) of the Treaty on the Functioning of the European Union to State aid in the form of public service compensation granted to certain undertakings entrusted with the operation of services of general economic interest (notified under document C(2011) 9380) [2012] OJ L7/3–10, and (3) Communication from the Commission, European Union framework for State aid in the form of public service compensation [2012] OJ C 8/15–22. On 25 April 2012 the Commission adopted, as the fourth and final pillar of the SGEI package, (4) Commission Regulation on the application of Articles 107 and 108 of the Treaty on the Functioning of the European Union to de minimis aid granted to undertakings providing services of general economic interest [2012] OJ L 114/8. 32 See sections 1.3 and 1.4 for an overview of these cases.

164

Capacity Mechanisms and State Aid Control

107(1) TFEU is an ‘effects-based’ test—so that the form and the objective of the national measure in question is irrelevant.33 The most problematic of these conditions in relation to capacity mechanisms are (a) whether the state measure in question confers a selective economic advantage on particular undertakings and (b) whether that advantage is funded by or through state resources. With respect to the remaining two conditions—distortion of competition and adverse effect on inter-state trade—it may be presumed that, as is generally the case, these conditions can be easily met in a liberalized market. This indeed is the Commission’s position in both the EEAG 2014–2020 and its earlier November 2013 Communication. This section considers the types of arguments which have been advanced to claim that a form of support to a sector does not amount to state aid at all.

9.3.2 The concept of economic advantage An important issue—and one that the EEAG remains silent upon—is whether undertakings entrusted with an obligation of providing reserve capacity in order to ensure generation adequacy and system reliability are being compensated for the performance of tasks that they would not have assumed under normal market conditions. If this is indeed the case, any state support might be viewed as mere compensation for the imposition of this obligation and cannot confer an economic advantage on the generator. However as the costs arising from regulatory obligations imposed by the state can in principle be considered to relate to the inherent costs of the economic activity, any compensation for these costs confers an economic advantage on the undertaking. This means that the existence of an economic advantage is in principle not excluded by the fact that the benefit does not go beyond compensation for a cost stemming from the imposition of a regulatory obligation.34

9.3.3 Compensation for PSOs In July 2003 the CJEU reversed its earlier—somewhat benign jurisprudence on the socalled ‘compensation approach’ to the provision of public services35 in the landmark Altmark case.36 As a result, in order to successfully claim that a particular form of compensation is not a state aid measure within the meaning of Article 107(1), and therefore is not subject to Commission approval, Member States must fulfil all four cumulative Altmark conditions to avoid notification of a capacity mechanism as a state aid within the meaning of Article 107(1) TFEU. First, the recipient undertaking must 33 See for example Case T-525/08 Poste Italiane (13 September 2013, nyr), and the jurisprudence cited therein. 34 See Communication from the Commission, Draft Commission Notice on the notion of State aid pursuant to Art 107(1) TFEU, issued for consultation on 17 January 2014, available at http://ec.europa.eu/ competition/consultations/2014_state_aid_notion/draft_guidance_en.pdf, accessed 1 February 2015, para 71. On compensation for stranded costs, see Case T-25/07 Iride SpA and Iride Energia SpA v Commission [2009] ECR II-245, paras 46 to 56. 35 Public service obligations (PSOs) may be imposed by the public authorities on the body providing a public service (airlines, road or rail carriers, energy producers and so on). 36 Case C-280/00 Altmark Trans [2003] ECR I-07747, paras 87–95.

9.3 Funding capacity mechanisms: When do state aid rules apply?

165

actually have valid public service obligations to discharge, and the obligations must be clearly defined. Second, the parameters on the basis of which the compensation is calculated must be established in advance in an objective and transparent manner. Third, the compensation cannot exceed what is necessary to cover all or part of the costs incurred in the discharge of public service obligations, taking into account the relevant receipts and a reasonable profit. Fourth, where the undertaking which is to discharge public service obligations is not chosen following a public procurement procedure to select a tenderer capable of providing those services at the least cost to the community, the level of compensation needed must be determined on the basis of an analysis of the costs which a typical undertaking, well-run and adequately provided with means to meet the public service requirements, would have incurred in discharging those obligations, taking into account the relevant receipts and a reasonable profit for discharging the obligations. The Commission has given further guidance on the application of these onerous conditions in its Communication on the application of the EU state aid rules to compensation granted for the provision of SGEI.37 As discussed earlier, in section 1.3.2, a claim that a capacity mechanism is a form of PSO—and not a form of state aid at all—may well be an attractive and feasible approach for Member States. At the same time past precedent has confirmed that in general, it is not always easy for a Member State to satisfy the fourth Altmark condition—that is that the undertakings in question must be selected by way of an open tender procedure.38 As a number of national capacity mechanisms examined in chapters 12 to 22 illustrate these schemes are often based on auction procedures. It should however be stressed that the Commission will often also scrutinize tender procedures to ensure that they do not merely involve competition for a subsidy. This then leaves the key question of whether the first Altmark test can be satisfied—that is are Member States entitled to entrust PSOs (or SGEIs) to electricity generators? As already noted, the Commission initially took the view in its first (leaked) draft of the EEAG from July 201339 that this was not the case. The November 2013 Communication announced a change of heart, and as already noted, the draft EEAG published one month later on 18 December 2013, confirmed that Member States would indeed be entitled to consider that certain services in the energy sector should be regarded as SGEI.40 This change of approach on the part of the Commission is hardly surprising. The right of Member States to designate services as SGEIs is enshrined in the Treaty, in Articles 1441 and 106(2), as well as 37

Communication from the Commission (n 34) p 4. As to the alternative test, see Case SA. 20350 State aid implemented by the Czech Republic for several regional bus service operators in the Ústí Region (Czech buses) [2014] OJ L 329/35. 39 Draft EEAG 2014–2020, non-published version of July 2013 (n 28). 40 Contrast the Commission decision of 20 November 2013 in Case SA.36740 AB Klaipėdos nafta (Lithuania LNG Terminal) (nyr), with the Commission decision of 18 December 2013 to initiate the formal investigation procedure in Case SA.34947 (2013/C)(ex 2013/N) United Kingdom Investment Contract (early Contract for Difference) for the Hinkley Point C New Nuclear Power Station (Hinkley Point) [2014] OJ C/69. 41 Article 14 TFEU requires the Union and the Member States, without prejudice to Arts 93, 106, and 107 TFEU, to use their respective powers in such a way as to make sure that services of general economic interest operate on the basis of principles and conditions, particularly economic and financial conditions, 38

166

Capacity Mechanisms and State Aid Control

Protocol 26 on SGI,42 and could hardly be eliminated let alone modified by a set of soft-law guidelines. Even if Article 3(2) of the 2009 Electricity and Gas Directives43 purport to harmonize the grounds on which Member States can invoke Article 106(2) TFEU, the CJEU has recognized that Member States may still invoke national policy consideration to justify departures from the Directives’ provisions.44 In this connection, the Commission has also held in its decision on state aid to an LNG terminal in Lithuania that Union rules consider security of supply as an objective that might justify public service obligations in accordance with Article 3(2) of the 2009 Gas Directive.45 The final version of the EEAG 2014–2020 is silent on the application of Article 106(2) TFEU but the section on generation adequacy (see below at section 9.4) recognizes that market and regulatory failures may cause insufficient investment in generation capacity.46 As a result, and on the assumption that all four of the Altmark conditions can be satisfied, it is possible for Member States to maintain that a specific capacity mechanism does not confer an economic advantage on the undertakings concerned but is instead compensation for a genuine SGEI or PSO. Alternatively, as discussed below in section 9.5, even if these conditions are not all met, a Member State can still invoke Article 106(2) of the Treaty as a legal basis for compatible state aid.47

9.3.4 Financed by or through state resources On the assumption that the Altmark tests are either not invoked by a Member State as an argument that the capacity mechanism in question is not a form of state aid, in the first place, or alternatively that any of the four conditions are not met, the only ‘escape-route’ open to a Member State wishing to avoid the state aid discipline altogether will be to argue that the measure or scheme in question is not financed by or through state resources and therefore notification is not required. This strategy may

which enable them to fulfil their missions. For certain services of general economic interest to fulfil their mission, financial support from the state may prove necessary to cover some or all of the specific costs resulting from the public service obligations. 42 Protocol (No 26) on Services of General Interest, TFEU. 43 Directive 2009/72/EC of the European Parliament and of the Council of 13 July 2009 concerning common rules for the internal market in electricity and repealing Directive 2003/54/EC [2009] OJ L 211/55 (2009 Electricity Directive); Directive 2009/73/EC of the European Parliament and of the Council of 13 July 2009 concerning common rules for the internal market in natural gas and repealing Directive 2003/55/EC [2009] L 211/94 (2009 Gas Directive). 44 See Case C-265/08 Federutility and others v Autorità per l’energia elettrica e il gas [2010] ECR I-3377, paras 35–39: ‘First, such an intervention must be limited in duration to what is strictly necessary in order to achieve its objective [ . . . ]. Secondly, the method of intervention used must not go beyond what is necessary to achieve the objective which is being pursued in the general economic interest. [ . . . ] Thirdly, the requirement of proportionality must also be assessed with regard to the scope ratione personae of the measure, and, more particularly, its beneficiaries.’ 45 2009 Gas Directive (n 43). Lithuania LNG Terminal (n 40) para 203. For a different approach, see Hinkley Point (n 40). 46 EEAG 2014–2020 (n 4) para 218. 47 Also discussed in sections 1.3 and 1.5 above. See also Communication from the Commission, European Union framework (n 31).

9.3 Funding capacity mechanisms: When do state aid rules apply?

167

too prove difficult, given the courts’ willingness to deem this condition to be satisfied even for indirect funding. The European courts have consistently interpreted Article 107(1) TFEU as an effectbased test: the form of state support is not relevant for the application of Article 107(1) TFEU.48 Following the seminal ruling in Case C-379/98 PreussenElektra49 the Court clarified that the state measure in question must be funded by or otherwise linked to ‘state resources’ in order to fall within the scope of Article 107(1) TFEU. It rejected the Commission’s request to extend the scope of the Article to ‘measures having equivalent effect to a state aid.’ Only advantages granted directly or indirectly through state resources can constitute state aid within the meaning of Article 107(1) TFEU. As the Advocate General (AG) in Case C-262/12 Vent de Colère50 recalled ‘Article 107 TFEU includes all the financial resources which the State may in fact use to support undertakings. That those resources constantly remain under public control, and therefore available to the competent national authorities, is sufficient for them to be categorised as state resources and for the measure to fall within the scope of Article 107(1) TFEU.’ Most capacity mechanisms are not financed directly by the State through direct grants or subsidies to generators, but by means of a levy or supplement imposed on electricity suppliers (or customers) and so the key question is whether these types of national mechanisms can be considered to be advantages granted indirectly through state resources. As the Court confirmed in PreussenElektra, where final consumers assume the responsibility for paying for the extra costs associated with supporting RES, Article 107(1) TFEU does not apply. If however consumers are required to pay extra charges or levies or a tariff component designated to finance a particular state-imposed objective, and the state assumes control over the collection and subsequent disbursement of the proceeds of this type of mechanism, then the test of ‘indirect control’ can be met. The resources in question—even if private in origin—are under the control of the state.51 As part of SAM,52 the Commission produced a Notice on the notion of state aid in January 2014.53 This document is intended to offer clarification only. ‘Considering that the notion of State aid is an objective and legal concept defined directly by the Treaty, the Commission will simply clarify how it understands the Treaty provisions, in line with the EU case law, without prejudice to the interpretation of the Court of Justice of the European Union.’54 Its section 3.2 sets out the Commission’s somewhat strict standpoint as to when state resources can be considered to be mobilized in support of a state measure: ‘The origin of the resources is not relevant provided that, before being directly or indirectly transferred to the beneficiaries, they enter under public

48 49 50 51 52 53 54

Case 173/73 Italy v Commission [1974] ECR 709, para 13. Case C-379/98 PreussenElektra AG v Schhleswag AG [2001] ECR I-2099. Case C-262/12 Association Vent de Colère! and others (19 December 2013, nyr) para 34. Case 76/78 Steinike & Wenlig v Germany [1977] ECR 595. SAM (n 22). Communication from the Commission (n 34). Communication from the Commission (n 34) para 3.

168

Capacity Mechanisms and State Aid Control

control and are therefore available to the national authorities, even if the resources do not become the property of the public authority.’55 Thus, in the Commission’s view, subsidies financed through parafiscal charges, levies, or compulsory contributions imposed by the state and managed and apportioned in accordance with the provisions of public rules imply a transfer of state resources, even if not administered by the public authorities.56 Moreover, the mere fact that the subsidies are financed in part by voluntary private contributions is not sufficient to rule out the presence of state resources, since the relevant factor is not the origin of the resources, but the degree of intervention of the public authority within the definition of the measure and its method of financing.57 For example, in its recent decision on the UK capacity mechanism58 the Commission found that the system of financing the relevant scheme as notified by the UK government concerns state resources. The UK will set up a state-owned Settlement Body to supervise the collection of the surcharge or levy to be imposed on all licensed suppliers and it will supervise the settlement process and payments disbursed to the capacity providers.59 This classification was not disputed by the UK authorities. By way of contrast, the LNG supplement imposed through the national gas transmission tariff in Lithuania to fund the LNG terminal was initially claimed by the Lithuanian authorities to fall within the scope of PreussenElektra, and should not be considered state aid. However, as the levy was collected by the national TSO and applied to de-fraying the costs of the company responsible for constructing and operating the LNG terminal, the Commission classified the LNG supplement as a form of state aid.60 It is therefore not always the case that the Member State accepts that a measure is indirectly financed through state resources but on the basis of established precedent, it appears that the transfer of state resources can only be ruled out in very specific circumstances. For example where resources from the members of a trade association are earmarked for a specific purpose in the interest of the members, decided on by a private organization and with a purely commercial purpose, and as long as the Member State in question is simply acting as a vehicle in order to make the contribution introduced by the inter-trade organization compulsory.61

55 The Commission relies here on Case T-358/94 Air France v Commission [1996] ECR II-2109, paras 65 to 67, concerning an aid granted by the Caisse des Dépôts et Consignations which was financed with voluntary deposits of private citizens which could be withdrawn at any time. See also Case C-83/98 P France v Ladbroke Racing and Commission [2000] ECR I-3271, para 50. 56 Case 173/73 Italy v Commission [1974] ECR 709, para 16; Joined Cases C-78/90 to C-83/90 Compagnie Commerciale de l’Ouest [1992] ECR I-1847, para 35. C-206/06 Essent Netwerk Noord v Aluminium Delfzijl [2008] ECR I-5497, paras 58–74. 57 Joined Cases T-139/09, T-243/09, and T-328/09 France et al v Commission (judgment of 27 September 2012, nyr), paras 63 and 64. 58 Commission decision of 23 July 2014 in Case SA.35980 (2014/N-2) United Kingdom Electricity Market Reform—Capacity Market C(2014) 5083 final, [2014] OJ C/348 (UK capacity mechanism). 59 UK capacity mechanism (n 58) paras 109–111. 60 Lithuania LNG Terminal (n 40) paras 90–96. 61 See Case C-345/02 Pearle [2004] ECR I-7139, para 41 and Case C-677/11 Doux élevages SNC et al (judgment of 30 May 2013, nyr).

9.3 Funding capacity mechanisms: When do state aid rules apply?

169

In this context, the Commission’s draft Notice on the notion of state aid provides the specific example of ‘surcharges imposed by law on private persons’ and indicates that these can be qualified as state resources:62 This is the case even where a private company is appointed by law to collect such charges on behalf of the State and to channel them to the beneficiaries, without allowing the collecting company to use the proceeds from the charges for purposes other than those provided for by the law. In this case, the sums in question remain under public control and are therefore available to the national authorities, which is sufficient reason for them to be considered State resources.63 Since this principle applies both to public bodies and private undertakings appointed to collect the charges and process the payments, changing the status of the intermediary from a public to a private entity has no relevance for the State resources criterion if the State continues to strictly monitor that entity.64

It is therefore apparent that the relevant factor is not the origin of the resources but the degree of intervention of the public authority within the definition of the measure and its method of financing.65 At its paras 63–64 the Commission’s draft Notice on the notion of state aid indicates that regulation that leads to financial redistribution from one private entity to another without any further involvement of the state does not entail a transfer of state resources, but only if the money flows directly from one private entity to another, without passing through a public or private body designated by the state to administer the transfer. This is a narrow approach. The PreussenElektra case law described in paras 63–64 of the draft Notice concerns situations where there was no publicly controlled fund or mechanism but rather a direct flow from one private entity to another pursuant to regulation with no further involvement from the State. In its final decision on the German Renewables Energy Act (EEG 2012),66 the Commission seems to consider that where a state sets rules under which private operators collect and administer a levy or surcharge then this is sufficient to characterize the measure as aid—even if no publicly controlled intermediary fund has been set up.67 Shortly prior to the publication of the draft Notice on the notion of state aid, the CJEU had confirmed the application of Article 107(1) in Vent de Colère68 to the effect that a mechanism for offsetting in full the additional costs imposed on undertakings because of an obligation to purchase wind-generated electricity at a price higher than the market price that is financed by final consumers, must still be regarded as an intervention by the state or through state resources within the 62

Draft Commission Notice on the notion of State aid (n 53) para 66. Essent Netwerk Noord (n 56) paras 69–75. 64 Commission decision 2011/528/EU on state aid C 24/09 (ex NN 446/08) Austria—Green Electricity Act, OJ L 235, 10 September 2011, p 42, recital 76. 65 Commission decision of 25 November 2014 in Case SA.33995 (2013/C) (ex 2013/NN) Germany— Support for renewable electricity and reduced EEG-surcharge for energy-intensive users its decision to open an in-depth investigation (C(2014) 8786 final (EEG-surcharge). This decision is now under appeal: action brought on 28 February 2014 (Case T-134/14 Germany v Commission). See also the Commission Decision in Case N94/2010 UK feed-in tariffs, 14 April 2010, paras 65–69.6. 66 Erneuerbare-Energien-Gesetz, 21 July 2014 (BGBl. I S. 1066), last amended by BGBl. I S. 2406 (EEG). 67 European Commission, Press release IP 14/2122, 25 November 2014. 68 Vent de Colère (n 50). 63

170

Capacity Mechanisms and State Aid Control

meaning of Article 107(1) TFEU. In this case the sums intended to offset the additional costs arising from the obligation to purchase imposed on the undertakings are collected from all final consumers of electricity in France and entrusted to the Caisse des Dépôts et Consignations (CDC)—a public body—under a mandate from the French state, that provides administrative, financial, and accounting management services for the national regulator (Commission de Régulation de l’Énergie—CRE). The CDC also determines late payments or defaults in payment by final consumers and reports them to that regulatory authority. The Court held that, ‘the sums thus managed by the [CDC] must be regarded as remaining under public control.’69 All these factors were taken to distinguish the present case from the set of facts at issue in PreussenElektra: Consequently, the funds at issue [in PreussenElektra] could not be considered a State resource since they were not at any time under public control and there was no mechanism, such as the one at issue in the main proceedings in the present case, established and regulated by the Member State, for offsetting the additional costs arising from that obligation to purchase and through which the State offered those private operators the certain prospect that the additional costs would be covered in full.70

On 11 December 2014, the General Court handed down its judgment dismissing an appeal by Austria against a Commission decision that a provision of the revised Austrian Green Electricity Act would result in the imposition of extra costs on undertakings that do not qualify for the relevant exemption, and so constituted unlawful state aid. The General Court upheld the Commission’s finding that the partial exemption constituted state aid, and rejected arguments advanced by Austria, supported by the UK, that the mechanisms at issue were similar to the legal framework underpinning the German feed-in tariff system in the PreussenElektra case. Given that the funds at issue were under constant state control, the General Court confirmed that in the light of the tests laid down in the Essent Netwerk Noord case,71 the Austrian legislation involved the use of state resources (Case T-251/11 Austria v Commission).72

9.3.5 Conclusion In the light of this recent case law it will be a challenge for Member States to set up a capacity mechanism that escapes Article 107(1) TFEU, as in the PreussenElektra case. The latter case remains good law, as confirmed by the General Court in the Austria v Commission case, but it would seem that the ‘control’ test is applied strictly. In practice, as confirmed by the recent Commission decision on the German EEG 2012,73 this would mean that the scheme must entail no public intervention in any form whatsoever by means of management of the private or public entities entrusted to collect a surcharge or levy. In so far as an intermediate fund—whether this is managed by a TSO or a specific settlement body—is created to collect surcharges or levies from final consumers then, in the light of recent case law and providing the remaining conditions 69 71 73

Vent de Colère (n 50) para 33. Essent Netwerk Noord (n 56). EEG-surcharge (n 65).

70 72

PreussenElektra (n 49) para 36. Case T-251/11 Austria v Commission [2015] OJ C 46/42.

9.4 Compatible support—the new EEAG 2014–2020

171

of Article 107(1) TFEU are met, it is highly unlikely that the scheme or measure in question cannot qualify as a form of state aid.

9.4 Compatible support—the new EEAG 2014–2020 The new EEAG 2014–2020 have replaced the EAG 200874 and came into effect on 1 July 2014. The EEAG is the first comprehensive set of guidelines to include guidance on aid to the energy sector beyond aid to support renewable energy and energy efficiency. Prior to the adoption in April 2014 of the EEAG, the only energyspecific guidance related to the methodology for analysing state aid linked to stranded costs in the electricity sector (the Stranded Costs Communication, discussed in section 9.2).75 The decision to include aid to promote certain objectives in the energy sector in the new guidelines was not without controversy. The Commission commenced its public consultation in July 2012 with a questionnaire on the EAG 2008, followed by consultation on an ‘issues paper’ of March 2013.76 A workshop was then held in order to consider the responses to this paper in April 2013. These events were then followed by a leaked draft version of the proposed guidelines, which included a specific set of assessment criteria in relation to nuclear power.77 This version was withdrawn following considerable internal and external criticism and an official draft was published for consultation on 18 December 2013.78 The text of the leaked draft of July 2013 suggested that the Commission took as its starting point the view that aid to promote ‘capacity mechanisms’ as they were originally referred to, would have to be carefully controlled in order not to unduly hamper the development of the single market in electricity. The final version published in April 2014, building on the November 2013 Communication,79 and the draft version of the EEAG issued for consultation on 18 December 2013, acknowledges that the shift to a RES-based system may necessitate state intervention and support to ensure the availability of generation capacity, albeit that it would be for the Commission to ensure that such mechanisms are designed so as not to unduly distort competition or hamper the development of the internal electricity market.

74

EAG 2008 (n 21). Stranded Costs Communication (n 11). See further Hancher et al, EU State Aids, 4th edn (Sweet & Maxwell, 2012) ch 19. 76 European Commission, Environmental and Energy Aid Guidelines 2014–2020, Consultation Paper, 11 March 2013. This paper indicated at p 3 that the new guidelines would address capacity mechanisms: ‘The objective of system stability is a legitimate concern of Member States. However, DG COMP is considering how best to ensure that state aid is restricted to situations where markets are not able to deliver the necessary generation capacity. Once a market failure is established, it needs to be demonstrated that State aid is an appropriate means to ensure system stability provided that alternative measures such as better interconnection, demand response or energy savings could not alleviate such concerns. If State aid is an appropriate instrument, it needs to be considered what compatibility conditions are necessary to prevent harm to the internal energy market by nationally focused measures.’ 77 Draft EEAG 2014–2020, non-published version of July 2013 (n 28). 78 Draft EEAG 2014–2020, version of 18 December 2013 (n 30). 79 November 2013 Communication (n 3). 75

172

Capacity Mechanisms and State Aid Control

The EEAG as finally adopted, contains a separate section on ‘aid for generation adequacy’ (section 3.9) as well as a section on aid to energy infrastructure (section 3.8). The criteria for assessment of aid schemes and individual aid measures are now more carefully spelt out compared to the corresponding sections in earlier drafts. In order to be declared compatible aid to support generation adequacy the national state aid measure or scheme must comply with the common assessment principles— section 3.1 and 3.2 as further tailored to capacity mechanisms in the specific conditions set out in section 3.9. Where the aid is below €15 million per project per undertaking, then it is not notifiable and will be considered compatible aid as long as it complies with the General Block Exemption Regulation (GBER) 2014.80 Section 1 of the EEAG includes a number of definitions pertaining to capacity mechanisms. In particular ‘generation adequacy’ means a level of generated capacity which is deemed to be adequate to meet demand levels in the Member State in any given period based on the use of a conventional statistical indicator used by organizations which the Union institutions recognize as performing an essential role in the creation of a single market for electricity, such as ENTSO-E.81 and a ‘generation adequacy measure’ means ‘a mechanism which has the aim of ensuring that certain generation adequacy levels are met at national level’.82 These definitions would appear to encompass a wide variety of measures, and the only constraining factor would seem to be that the demand levels should be based on harmonized statistics.

9.4.1 Common assessment principles The EEAG 2014–2020 sets out seven general or common assessment criteria, which must be met in any event for a state aid measure to promote generation adequacy to be declared compatible. These general compatibility conditions apply to all aid falling within the scope of the guidelines, unless the more specific sections of chapter 3 of the EEAG apply. According to the conditions, the measure must (a) contribute to a welldefined objective of common interest, (b) be targeted and necessary—ie, it must remedy a well-defined market failure, (c) be an appropriate policy instrument to address the common objective (d) have a proven incentive effect (e) be proportionate and limited to the minimum amount needed to incentivize additional investment, (f) its positive effects must outweigh the negative effects on competition and trade between the Member States, and finally (g) it must be transparent so that economic operators and the public must have easy access to relevant acts and pertinent information about the aid awarded. Section 3.9.1 of the EEAG on aid for generation adequacy further elaborates on when such a measure will be considered to contribute to an objective of common interest. 80 Commission Regulation (EU) 651/2014 of 17 June 2014 declaring certain categories of aid compatible with the internal market in application of Articles 107 and 108 of the Treaty [2014] OJ L 187/1. 81 EEAG 2014–2020 (n 4) para 35. See also section 3.9.1 at para 222: the identification of the generation adequacy problem should be consistent with the analysis carried out by ENTSO-E in accordance with its tasks under the 2009 Cross-border Regulation (n 19), in particular Art 8. 82 EEAG 2014–2020 (n 4) para 36. Broadly similar definitions of ‘generation adequacy’ and ‘capacity adequacy’ were annexed to the draft EEAG 2014–2020, version of 18 December 2013 (n 30).

9.4 Compatible support—the new EEAG 2014–2020

173

The compatibility criteria appear to leave ample scope for Member States to design the form of investment or operating aid83 it intends to confer and to pursue the objectives of addressing both short-term concerns relating to system adequacy caused by the lack of flexible generation and/or to define targets for generation adequacy ‘which Member States wish to ensure regardless of short term considerations’. Member States are encouraged but not obliged to avoid achieving generation adequacy by subsidizing environmentally or economically harmful forms of generation and to consider alternative strategies such as DSR and increasing interconnection capacity.84 Section 3.9.2 of the EEAG on the necessity of the state aid measure requires a clear analysis and quantification of the nature and causes of the generation adequacy problem. The national authorities should also clearly demonstrate the reasons why the market cannot deliver adequate capacity in the absence of intervention, taking into account ongoing market and technology developments, including the development of market coupling, intraday markets, balancing and ancillary service markets, and storage. Emphasis will be placed in the Commission’s own assessment of the impact of variable generation from other Member States, demand side participation and measures to encourage demand side management, the potential contribution of actual and potential interconnectors and any other element which might cause or exacerbate generation adequacy problems including regulatory failures such as caps on wholesale prices. In other words, the Member State must demonstrate that alternative means to address generation adequacy cannot meet the required targets. Should the Commission consider that the cause of the problem lies in regulated wholesale pricing—which is no longer allowed in accordance with the EU’s internal electricity market legislation—it may presumably cast doubt on the necessity of the proposed aid measure.85 Section 3.9.3 of the EEAG on the appropriateness of aid makes it evident that the measures should only remunerate the service of pure availability and not for the sale of electricity. It should therefore not merely alleviate stranded costs or favour a particular technology. The measure should be open and provide adequate incentives to both existing and future generators as well as substitutable technologies, such as DSR and storage solutions. It should also be delivered to allow for potentially different lead times, corresponding to the time needed to realize new investments by new generators using different technologies. Once again, the extent to which interconnection capacity could remedy any possible problem of generation adequacy should be take into account.

83 ‘Operating aid’ means aid aimed to reduce an undertaking’s current expenditure that is not related to an initial investment. 84 Para 197 of the draft version of EEAG of 18 December 2013 was more strictly worded in that it required that Member States ‘should first take into account alternative ways of achieving this objective which do not negatively impact on the objective to phase out environmentally harmful subsidies’ albeit that it went on to recognize that a transitional period may be required in order to achieve the aim of ensuring sufficient generation adequacy and security of supply. See draft EEAG 2014–2020, version of 18 December 2013 (n 30) para 197. 85 The Commission has in fact launched infringement proceedings against a number of Member States who have not abolished wholesale pricing.

174

Capacity Mechanisms and State Aid Control

Section 3.9.4 of the EEAG requires that the incentive effect of the aid will be assessed on the basis of the common compatibility conditions set out at section 3.1.4. An incentive effect occurs when the aid induces the beneficiary to change its behaviour to improve the functioning of a secure, affordable, and sustainable energy market, a change in behaviour which would not occur without the aid. These compatibility conditions are sharpened in the new EEAG compared to previous environmental aid guidelines with the addition of a number of procedural requirements. The EEAG introduces stricter conditions in respect of establishing the so-called ‘incentive effect’ of a state aid measure for the beneficiary, and requires the submission of an application form.86 In order to satisfy the test of proportionality, section 3.9.5 of the EEAG requires that the overall amount of aid should allow beneficiaries a reasonable rate of return. A competitive bidding procedure that is based on clear, transparent, and nondiscriminatory criteria is considered as leading to reasonable rates of return. Overcompensation must be avoided and the measure should include mechanism to avoid windfall profits. The price paid for the availability must automatically tend to zero when the level of capacity supplied is expected to be adequate to meet the level of capacity demanded. Section 3.9.6 of the EEAG deals with the balancing test—the avoidance of negative effects on competition and trade. Positive elements include completive bidding and technological neutrality87 as well as the inclusion of DSR, interconnectors, and storage and the guarantee of participation by a sufficient number of generators to establish a competitive price for the capacity. The measure should not reduce incentives to invest in interconnection capacity, or undermine market coupling, including balancing markets. It should not ‘unduly strengthen market dominance’ and should give preference to low-carbon generators in case of equivalent technical and economic parameters. Importantly, the section indicates that the participation of operators from other Member States should be taken into account in so far that participation is physically possible and the capacity can be physically provided to the Member State implementing the measure and the obligations set out in the measure can be enforced. This would seem to allow the exclusion of operators from other Member States if the notifying government can establish that either sufficient capacity is not available in interconnectors, or that they have no means of sanctioning a beneficiary who subsequently reneges on its obligations to provide availability. It is well established in the case law that a state aid measure must not contain conditions—including its financing method—which entail a non-severable violation of Union law. Thus the measure must not be in breach of Articles 30 and 110 TFEU or indeed Articles 34 and 36 TFEU.88 This topic will be discussed in greater detail below, in section 9.5.2.

86

EEAG 2014–2020 (n 4) paras 49–52. The draft version of December 2013 contained a more explicit set of requirements with respect to the technological neutrality of the measure. See draft EEAG 2014–2020, version of 18 December 2013 (n 30) para 212. 88 See Hancher et al (n 75) chapter 3. 87

9.5 Assessment

175

Unlike many other categories of aid covered by the EEAG, aid for generation adequacy is not subject to additional requirements on aid intensity and so it may be assumed that the maximum of 100 per cent coverage is allowed.89 Finally, it is important to note that section 5 of the EEAG 2014 introduces the possibility for the Commission to require that certain aid schemes are subject to a time limitation (usually four years or less) and to an ex post evaluation as referred to in para 29 of the Guidelines. Such evaluations are to be carried out for schemes where the potential distortion of competition is particularly high. This requirement is only to apply to aid schemes which large aid budgets, containing novel characteristics or when significant market, technology, or regulatory changes are foreseen. The evaluation must be carried out by an expert independent from the grant-giving authority on the basis of a common methodology provided by the Commission and must be made public. The Commission has now published the common methodology that it expects Member States to apply.90 The precise scope and modalities of each evaluation will be defined in the decision approving the aid scheme. Any subsequent aid measure with a similar objective must take into account the results of the evaluation. In its decision approving the UK capacity mechanism,91 discussed below in section 9.5, the Commission did not include a requirement for an ex post evaluation report.

9.5 Assessment All measures notified to the Commission in respect of which it is called upon to take a decision after 1 July 2014 are now assessed under the new EEAG 2014–2020. Hence measures notified prior to July 2014 but not yet ruled upon will be subject to these new guidelines. To date, only one decision is publicly available—the decision on the UK capacity mechanism.92 It is understood that several other national schemes are under consideration. It is therefore somewhat premature to offer any meaningful assessment of the EEAG 2014 and its application in practice. Several criteria are of relevance in the light of the wider goals of the SAM exercise, of which the EEAG form part, that is, ensuring a coherent approach to assessment which allows faster decision making on the part of the Commission, while promoting efficient and targeted use of public resources on the part of the Member State. A first key question is therefore whether the new compatibility criteria are sufficiently clear and precise in order for Member States to design their support mechanisms accordingly, and thus obtain a positive decision without a prolonged period of negotiation with the Commission or, indeed, without the latter having to resort to a formal investigation. It will be recalled that this latter step is only required if the Commission has serious doubts as to the compatibility of the aid measure with

89

There is no reference to section 3.9 in the annex on aid intensities for investment aid. Commission Staff Working Document, Common methodology for State aid evaluation, SWD (2014) 179 final, 28 May 2014. 91 92 UK capacity mechanism (n 58). UK capacity mechanism (n 58). 90

176

Capacity Mechanisms and State Aid Control

Article 107(3).93 Member States may however avoid this if they are prepared to accept the necessary amendments—by agreeing to commitments—to remove any doubts that the Commission may entertain as to the compatibility of the (notified) measure. Interested parties, including potential aid beneficiaries and their competitors, have formal rights to submit observations in the investigation phase only, and may therefore welcome a more prolonged enquiry.94 As already mentioned, the EEAG 2014 does not deal with the complex issue of when a state support measure is to be classified as a form of state aid within the meaning of Article 107(1) TFEU nor does it give any additional guidance on whether such support, if financed by state resources, could be deemed to be compensation for the provision of a public service obligation. It is not unusual for a Member State to claim that (a) the measure is not funded through state resources and therefore not notifiable and (b) even if it is state financed, the measure is covered by the cumulative Altmark criteria and is also not notifiable, or (c) that in the alternative, it is compatible state aid on the basis of the Guidelines or on the basis of Article 107(3), or in the alternative, (d) that the measure is compatible with Article 106(2) TFEU and the related SGEI Framework of 2011. This means that, in practice, the application of the relevant compatibility conditions in the Guidelines is often only one part of the assessment process. The Commission will be required to assess all four sets of arguments, where these have been advanced by the Member State granting the support. This can lead to lengthy and detailed ‘opening decisions’ to launch the formal investigation phase. Even if the Commission considers the aid to be compatible after its preliminary investigation, its decision closing the procedure and finding the measure to be either compatible aid or no aid at all, can also run into many pages. For example, in the Lithuania LNG Terminal case95 the Commission’s decision to declare the aid compatible at the end of the first or preliminary investigation runs to fifty-one pages. This was perhaps in part due to the fact that the case concerned a package of measures which included both investment and operating aid. The Lithuanian LNG decision is by no means unique. The Commission’s decision to open a formal investigation into the German support for RES and a reduced EEGsurcharge for energy intensive users (EEG-surcharge) totals fifty-seven pages.96 The decision to open a formal investigation into nuclear aid for Hinkley Point ran to eighty pages.97 Nevertheless, the availability of clearer guidelines on the application of Article 107(3) TFEU and the promise of a fast decision may deter national authorities from deploying all possible lines of argument to defend the measure at hand. Indeed, in two of the three cases mentioned earlier no specific guidelines were in fact applicable.

93 Article 4(4) of Council Regulation 659/1999 of 22 March 1999 laying down detailed rules for the application of Article 108 of the Treaty on the Functioning of the European Union of the EC Treaty [1999] OJ L 83/1, as amended. 94 It is not uncommon for submissions to be made during the first phase—see in this respect the various submissions examined in some detail in the Commission decision on the UK capacity mechanism. See UK capacity mechanism (n 58) paras 2.11–2.13. 95 96 Lithuania LNG Terminal (n 40). EEG-surcharge (n 65). 97 Hinkley Point (n 40).

9.5 Assessment

177

A second and more important question is whether the new guidelines are sufficiently flexible to accommodate diverse national situations but, at the same time, can ensure an adequate level of coherence across the EU, so that the common objectives announced in the SAM exercise as well as the EEAG 2014–2020 can indeed be safeguarded, and in particular that the completion of the internal electricity market is indeed not endangered. One obvious problem facing the Commission at this early stage in its experience of applying the generation adequacy guidelines now anchored in the EEAG is that it must essentially rely on national assessments of generation adequacy given the absence of a harmonized EU methodology, database and scenarios.98 If the Member State asserts that it faces imminent security of supply threats, it is unlikely that the Commission will be in a sufficiently strong position to cast doubt on the national assessment of the necessity for support.99 This means that it is likely to focus its assessment on the appropriateness and proportionality of the national measures and the avoidance of negative effects on competition and trade.

9.5.1 The first application of the EEAG 2014 to capacity mechanisms As of yet there is only one available published decision—the Commission’s clearance of the UK capacity mechanism notified to it on 23 June 2014 and cleared on 24 July 2014, albeit following ‘pre-notification contacts’.100 The development of the UK capacity mechanism in the context of the UK’s Electricity Market Reform (EMR) exercise has been discussed in detail in chapter 22. The UK mechanism is based on capacity auctions.101 This chapter will only consider the Commission’s assessment of the UK scheme under section 3.9 of the EEAG 2014–2020. It may be noted that the Commission decision contains a useful summary of the notified measure at para 2, as well as a detailed description of it, running to some twenty pages. As further described at paras 61 to 68, the UK scheme follows a ‘delivered energy’ model: capacity providers are obliged to deliver energy whenever needed to ensure security of supply, ie in real system stress situations, and face penalties if they fail to do so. As the UK did not dispute that the notified measure was financed through state resources nor did it invoke the Altmark tests, the Commission was only required to assess the compatibility of the scheme on the basis of section 3.9 of the EEAG.102 With regard to the first two tests laid down in section 3.9 of the EEAG—that the measure contributes to an objective of common interest and is a necessary measure—the UK government identified two market failures that prevent the market from bringing the necessary capacity to meet the established generation adequacy standard.103 Further, the UK claimed that reliability is a public good. 98

Discussed at length in chapters 3 and 5. As indicated in section 9.2 of this chapter, the Commission is currently working on a common methodology. 100 UK capacity mechanism (n 58). 101 For an explanation of capacity auctions in general, see section 1.2.3.2. See further chapter 22 on the UK capacity auctions. As noted in section 1.2.3.4, the UK reform of the balancing system may lead to a market for reliability options (UK capacity mechanism (n 58) para 92). 102 UK capacity mechanism (n 58) para 117. 103 UK capacity mechanism (n 58) paras 83–86 and Table 1 at p 34. 99

178

Capacity Mechanisms and State Aid Control

The Commission considered the objective of common interest and necessity of the aid together. It considered that the UK had adequately demonstrated that capacity adequacy could reach critical levels as of 2018–2019, and this had also been confirmed by the latest ENTSO-E system adequacy report.104 In respect of the first capacity auction held on 16 December 2014, the UK government had followed the national TSO (National Grid) and taken a conservative estimate of the contribution from interconnector flows. Similarly, the potential contribution of DSR would be kept under review and transitional auction arrangements would be made to support the growth of DSR.105 In this respect the UK offered important commitments to re-assess the contribution of interconnectors during stress events and review the methodology accordingly as of 2015, and to adjust the amount of capacity procured in subsequent auctions if the contribution of interconnectors in the four-year ahead auction planned for 2014 proves to be under estimated.106 Two other concerns raised in written submissions by third parties are dealt with in this context: the fact that it would lead to support for fossil fuel generators and may also favour new generators at the expense of existing plants and DSR. The Commission found that the UK had sufficiently explored ways to mitigate the negative impacts that the notified measure may have on the objective of phasing out environmentally harmful subsidies, including the award for contracts for difference (CfDs) to support low carbon generators. As existing generators and DSR operators would not be faced with the same investment costs as new plant, longer contracts (fifteen years) were justified for new investment, and hence the measure was viewed as technology neutral.107 In the meantime, National Grid announced the results of the pre-qualification round revealing that a considerable volume of prequalify capacity would be supplied by new CCGTs and would be awarded fifteen-year capacity contracts if successful in the first auction.108 Furthermore several interested parties have lodged appeals for the annulment of the Commission’s decision before the General Court.109 Next, the Commission assessed the measure to be appropriate in the light of the criteria set out in section 3.9.3 of the EEAG. It concluded that the UK capacity mechanism would support the development of an active demand side and will support increased liquidity and competition in both the capacity and the electricity markets. Smaller generators and DSR participants and suppliers have a clear route to market and receive a fair value for the capacity they provide. Despite somewhat tortuous reasoning, the Commission was able to conclude that the measure remunerates only the service of

104

105 UK capacity mechanism (n 58) para 119. UK capacity mechanism (n 58) para 122. 107 UK capacity mechanism (n 58) para 124. UK capacity mechanism (n 58) para 129. 108 See for a summary, see Frontier Economics and LCP, ‘Capacity Market - winter 2018/19 auction: What can we learn from the pre-qualification results?’ (Policy Brief, October 2014) available at http://www.frontiereconomics.com/publications/22117/, accessed 1 February 2015. The final results of the auction are available at https://www.emrdeliverybody.com/Shared%20Documents/Final%20Auction%20Results%20Report_v3.pdf, accessed 1 February 2015. For a comment on the results, see Frontier Economics and LCP, ‘Review of the first GB capacity auction’ (Policy Brief, January 2015) available at https://www.frontier-economics.com/docu ments/2015/01/lcp-frontier-economics-review-first-gb-capacity-auction.pdf, accessed 1 February 2015. 109 Press release about an appeal lodged by Tempus Energy, available at http://www.tempusenergy. com/press-release-04122014.pdf, accessed 1 February 2015. 106

9.5 Assessment

179

pure availability of capacity as beneficiaries are compensated for the units they make available and not for the energy delivered, even although as explained at paras 61–68, the scheme is in fact based on a ‘delivered energy’ model. The Commission also considered whether it was defensible to exclude foreign capacity from the scheme in its early stages. It will be recalled that although the amount of interconnected capacity is included in the calculation of the amount of capacity to be procured, interconnectors would only be allowed to bid in directly in the auction process, as if they were generators, once the scheme was adjusted in time for the second auction. The Commission was satisfied with this approach as it was similarly convinced that the differential treatment of existing and new plants was not discriminatory, and indeed certain requirements imposed on existing as opposed to new plants were justified in order to mitigate market power.110 As the support would be granted on the basis of a competitive bidding process, the measure is assumed to meet the procedural criteria in relation to its incentive effect. The substantive tests were also met as the measure would induce a change of behaviour so that the beneficiaries make available the necessary amount of capacity to meet the reliability standards set by the UK.111 As a counterfactual the Commission appears to have accepted the UK’s various simulations of investment in generation up to 2030 without further detailed scrutiny. The Commission also determined the scheme to be proportional: capacity providers would receive fair compensation, allowing them to earn a reasonable rate of return, and meet the relevant criteria for the proper design of a competitive bidding process to avoid the risk of windfall profits. Despite the complaint from an existing operator that the scheme would result in more new plant than necessary and therefore more aid than necessary, the Commission concluded that the number of new plants would be limited to the minimum necessary as competitive existing plants would be likely to bid at lower prices.112 Unsurprisingly in performing the balancing test, the Commission found that the positive effects outweighed any potential negative outcomes. In particular, it appeared to be convinced on the basis of the evidence supplied to it that the business case for future interconnection would not be undermined, the scheme would not depress electricity prices in the energy market, and new investment will be encouraged, countering the risk of market dominance from existing plant as price takers. The measure would also give preference to low-carbon generators, especially when assessed in conjunction with the UK’s Carbon Price Floor scheme introduced in 2013.113

9.5.2 Articles 30 and 110 TFEU Finally the Commission considered whether the financing mechanisms for the notified measure would infringe Articles 30 or 110 TFEU.114 As explained in greater detail in 110

111 UK capacity mechanism (n 58) paras 130–139. UK capacity mechanism (n 58) para 143. 113 UK capacity mechanism (n 58) para 147. UK capacity mechanism (n 58) para 153. 114 As indicated at para 29 of the EEAG 2014–2020 (n 135), if a state aid measure or the conditions attached to it (including its financing method when it forms an integral part of it) entail a ‘non-severable 112

180

Capacity Mechanisms and State Aid Control

chapter 11, the latter article forbids the imposition of any form of internal taxation or levy on imported products in excess of that imposed on similar domestic products whereas the former forbids the imposition of a charge on imported goods having equivalent effect to a customs duty.115 The CJEU has considered that a levy imposed on both imported and domestic electricity the proceeds of which were used to defray the stranded costs of domestic producers, could infringe these provisions.116 It should be recalled that the capacity payments would be financed by a levy imposed on all licensed electricity suppliers in relation to their market share based on electricity volumes sold. The Commission considers this levy to be equivalent to a tax on electricity consumed and given that interconnectors are excluded from the first auction, this levy would discriminate against imports. However, somewhat surprisingly, if not circuitously, the Commission considered that imported electricity was not in a similar situation to domestic electricity as the current constraints described earlier in its decision117 imply that ensuring generation adequacy at national levels can only be provided by domestic capacity. Hence the differential treatment was based on objective factors as required by Union law. It is noteworthy that in EEG-surcharge the Commission took a stricter approach and found that the ‘EEG-Act 2012 may prima facie in particular have a discriminatory effect in that §39 EEG-Act 2012 provides for a reduced rate of the EEG surcharge in case of so-called direct marketing that seems to be available only when the supplier has purchased 50% of his electricity portfolio from national RES electricity producers and seems therefore to constitute a discriminatory charge within the meaning of Article 110 TFEU.’118 It would appear that the Commission has now accepted a commitment from the German government to remedy the risk that the EEG 2012 would discriminate against imported electricity so that Germany would invest some €50 million in interconnectors and EU energy projects.119 Further, in the decision on the various CfDs concluded in the UK120 and in the related decisions on aid for the benefit of certain wind farm projects,121 as a result of the Commission’s concerns that this levy would indeed infringe Articles 30 and 110 TFEU, the UK was required to offer certain commitments to ensure that the levy to finance the CfDs would not discriminate against imported energy.

violation of Union law’, the aid cannot be declared compatible with the internal market. In the field of energy, any levy that has the aim of financing a state aid measure needs to comply in particular with Arts 30 and 110 TFEU. 115 See further Case C 213/95 Outokumpu Oy [1998] ECR 1–1777. 116 Essent Netwerk Noord (n 56) para 57. 117 UK capacity mechanism (n 58) para 20. 118 EEG-surcharge (n 65) para 247. 119 European Commission, Press release IP 14/2122 (n 67). 120 Commission decision of 23 July 2014 in Case SA.36196 (2014/N) United Kingdom Electricity Market Reform—Contract for Difference for Renewables, C(2014) 5079 final [2014] OJ C 393, para 87 ff. 121 Commission decision of 23 July 2014 in Cases SA.38758 (2014/N), SA.38759 (2014/N), SA.38761 (2014/N), SA.38763 (2014/N) & SA.38812 (2014/N) United Kingdom Support for five Offshore Wind Farms: Walney, Dudgeon, Hornsea, Burbo Bank and Beatrice [2014] OJ C 393, paras 80–86.

9.6 Conclusion

181

9.6 Conclusion In the absence of any other suitable or available weapons the Commission is left with the choice of pursuing infringement actions in the event, for example, that capacity mechanisms breach the EU rules on free movement and/or mobilizing the EU state aid rules. On the face of it the latter strategy has its attractions. Member States must notify their plans to grant aid and are bound by the ‘standstill’ provision in Article 108(2) TFEU to await a positive assessment of those plans by the Commission before putting them into effect. In order to obtain that approval the Member States may have to make concessions and commitments—and thus convince the Commission that the schemes in question fit the compatibility criteria developed at section 3.9 of the EEAG 2014. The Commission remains optimistic that this strategy will deliver the desired results. Indeed, in its recent Communication on Progress towards completing the Internal Energy Market, published in October 2014, the Commission claims that: Progress has also been made in addressing the threat of uncoordinated and countereffective national measures damaging the internal market. [ . . . ] In the State aid guidelines for energy and environment the Commission provided guidance to Member States to ensure that their interventions are necessary and proportionate pointing at their pivotal role in making the internal market a success rather than intentionally or unintentionally damaging it. [ . . . ] As a minimum requirement, the Commission asks for capacity mechanisms to be open to capacity abroad which can effectively contribute to meeting the required security of supply standards in the Member State concerned. A second requirement is that the capacity mechanisms must promote and reward demand side solutions to the same extent as generation solutions. Flexibility of production and demand must be encouraged so that capacity mechanisms complement in this respect the incentive stemming from variable electricity prices in the dayahead, intraday and balancing market.122

The UK capacity mechanism case analysed here suggests that the Commission will be satisfied that capacity mechanisms are compatible forms of state aid as long as there is hope of progress towards including interconnectors and DSR in the national market design in question. The real test for the effectiveness of the state aid guidelines will be whether the Member States deliver on their commitments, and whether the Commission can and will take action if they do not.

122

The October 2014 Communication (n 20).

10 Antitrust Law A Missing Piece in a Regulatory Puzzle? Adrien de Hauteclocque and Małgorzata Sadowska1

10.1 Introduction Capacity mechanisms are likely to play a central role in the future market design of most Member States. As expected, the current debate has focused on the wide variety of possible mechanism designs, and the distortions they would create in the internal energy market.2 The new subsidy schemes imply considerable public intervention at national level, prompting serious debate on the issue of EU state aid law.3 Indeed, the European interest in the implementation of capacity mechanisms appears to depend, almost entirely, on EU state aid control, which attempts to regulate ex ante the uncoordinated, if not chaotic, process. In July 2014, the Commission approved the UK capacity mechanism as compatible state aid.4 However, as of yet, the role that antitrust policy can play ex post has largely been neglected. It is no secret that EU energy companies have an impressive record of gaming the slowly liberalizing markets and exploiting new regulation. A capacity mechanism design, which is market-based, can affect incentives, and thus competition among firms, both inside and outside of the mechanism. In addition, capacity mechanisms will be difficult to amend ex post, suggesting a degree of path dependency. Therefore, it is critical to anticipate how the design will affect the behaviour of firms and explore whether the enforcement of antitrust rules can constitute a suitable fix. If this is not the case, due to the inherent limitations of antitrust laws, are there alternative legal tools, like the REMIT Regulation,5 which could provide an ex post control of the firms’ conduct? In essence, the choice of capacity mechanism design involves the standard trade-off between setting prices or quantities administratively, and a free market approach. Although superior in theory, a market-based design could facilitate anticompetitive behaviour. Of course, capacity mechanisms might distort the EU internal market in different ways, and only some of these issues can be addressed through the ex post control of the firms’ competitive behaviour, however there is little doubt that, for The authors would like to thank Lars Kjlbye for his helpful comments on the first draft of this chapter. For an overview of capacity mechanism types, see section 1.2.3. For discussions on possible market distortions, see, for example, sections 2.3, 3.3, 4.5, or 5.3.2. 3 See chapter 9. 4 Commission decision of 23 July 2014 in Case SA.35980 (2014/N-2) United Kingdom Electricity Market Reform—Capacity Market C(2014) 5083 final, [2014] OJ C/348 (UK capacity mechanism). For a discussion, see above, section 9.5.1. 5 Regulation (EU) No 1227/2011 of the European Parliament and of the Council of 25 October 2011 on wholesale energy market integrity and transparency [2011] OJ L 326/1-16 (REMIT). 1 2

10.2 EU enforcement in the energy sector—a primer

183

instance, a centralized capacity auction will make it much harder for verticallyintegrated incumbents to abuse their dominant position by favouring their subsidiaries, as noted in section 7.3.1.1 above. Policy makers should thus keep in mind that the capacity mechanism design ought to consider the pre-existing market structure. This chapter is a first attempt at screening and systemically classifying antitrust issues arising in the context of capacity mechanisms. It focuses on the issues that could reasonably be tackled under Article 101 TFEU (cartels and coordinated practices), Article 102 TFEU (the abuse of a dominant position), and parallel national provisions. It is worth mentioning that the available material on these issues is scarce to none, with the notable exception of certain answers to the Commission’s public consultation on generation adequacy,6 and some arguments that were raised in the Commission’s analysis of the UK capacity mechanism.7 Section 10.2 provides an overview of the legal and institutional framework, and draws on previous cases of antitrust enforcement in the energy sector. Section 10.3 presents a typology of likely antitrust issues depending on the capacity mechanism design. Section 10.4 provides some thoughts on market definition for the purpose of anticompetitive assessment. Section 10.5 covers selected anticompetitive practices that could be addressed under Articles 101 and 102 TFEU, and discusses the effectiveness of the antitrust response in these particular cases. Section 10.6 concludes.

10.2 EU antitrust enforcement in the energy sector—a primer As a general rule, EU antitrust laws apply to all aspects of the energy sector, as confirmed by the CJEU in the 1964 Costa v Enel case.8 The fact that EU sectorspecific rules exist does not limit their application. The CJEU also confirmed, in that case, that EU antitrust laws can be enforced even if a previous decision of a national regulator still leaves room for abuse. Since market liberalization, the Commission has largely applied the antitrust rules in this sector, in particular to supplement the slow development of EU sector-specific rules.9 The role of the Commission in the promotion of competition in the liberalized energy markets has evolved over time. Increasingly, it has resorted to quasi-regulatory measures to foster competition by way of antitrust enforcement. Accepting unilateral commitments by the firms involved in antitrust proceedings has also become the means to restructure the market and promote competition. The commitment procedure under Article 9 of Regulation 1/200310 allows the Commission to accept and make legally binding commitments offered by undertakings in the course of antitrust 6 European Commission, Consultation Paper on generation adequacy, capacity mechanisms and the internal market in electricity (the 2012 Consultation Paper). The Consultation Paper, together with all replies, is available at http://ec.europa.eu/energy/en/consultations/consultation-generation-adequacycapacity-mechanisms-and-internal-market-electricity, accessed 1 February 2015. 7 UK capacity mechanism (n 4). 8 Case 6/64 Costa v Enel [1964] ECR 585. 9 Hubertus von Rosenberg, ‘Unbundling through the back door . . . the case of network divestiture as a remedy in the energy sector’ (2009) European Competition Law Review 30 (5), 237–54. 10 Regulation 1/2003 of 16 December 2002 on the implementation of the rules on competition laid down in Arts 81 and 82 of the Treaty [2003] OJ L 1/1.

184

Antitrust Law: A Missing Piece in a Regulatory Puzzle?

proceedings, when it considers them sufficient to address the underlying competition problem.11 Article 9 has been inserted in order to promote procedural economy and speed. The discretion of the Commission when negotiating commitments is limited in theory by the suitability, necessity, and proportionality tests defined in Regulation 1/2003.12 They are intended to ensure that the Commission only addresses the underlying anticompetitive concerns, and does so effectively. The commitment procedure allows the Commission to negotiate liberalization outcomes directly with the incumbent, without going through the interface of national regulatory authorities and Member States. The tendency to use commitments instead of formal decisions has been largely criticized for its inability to clarify ‘the rules of the game’ in a newly opened sector with huge capital needs. Furthermore, the legitimacy of the Commission when it pushes forward an EU liberalization agenda through the antitrust rules has been questioned, as well as the suitability of the antitrust tool for this purpose.13 There are three periods of antitrust enforcement that are evident in the energy sector. A willingness to intensify the use of antitrust rules could be seen in the decision by the Commission to mount the energy sector inquiry, the conclusions of which were made public in 2007.14 The sector inquiry remains a useful source of information and has allowed the Commission to develop a more coherent plan for the subsequent antitrust enforcement. In the first phase, the Commission mainly focused on marketsharing agreements, with destination clauses and other territorial restrictions,15 and long-term supply contracts producing foreclosure effects.16 In the second phase, the Commission tackled more complex issues. Following the RWE17 and E.ON18 cases, EU antitrust enforcement was increasingly concerned with issues traditionally addressed through sector-specific regulation, in particular infringements linked to national networks, cross-border infrastructure, and price manipulation on the wholesale and balancing markets. Not surprisingly, these complex cases resulted in regulatory-like remedies. In the RWE case, under the antitrust rules, the Commission was tackling a fairly straightforward case of discriminatory (or inefficient) access to the network for competitors. But, in subsequent cases, the Commission demanded that dominant companies examine demand more effectively (ENI19), or even release capacity it was essentially using for itself (GDF20). In the Swedish Interconnectors case,21 the 11

Similar procedures exist at national level. Regulation 1/2003 (n 10). 13 Adrien De Hauteclocque, Market Building through Antitrust: Long-term Contract Regulation in EU Electricity Markets (Cheltenham: Edward Elgar, 2013). 14 Communication from the Commission, Inquiry pursuant to Art 17 of Regulation (EC) 1/2003 into the European gas and electricity sectors (COM/2006/851 final, January 2007, ‘Final Report’), accompanied by the DG Competition report on the energy sector inquiry (SEC(2006)1724, 10 January 2007, ‘DG COMP Report’). 15 See, for instance, Case COMP/37.811 Sonatrach, Case COMP/38.308 Gazprom/ENI, Case COMP/ 38.307 Gazprom/E.ON Ruhrgas, Case COMP/38.085 Gazprom/OMV and Case COMP/38.662 GDF/ENEL. 16 Case COMP/37.966 Distrigaz, Case COMP/39.387 Electrabel, Case COMP/39.386 EDF, and Case Exeltium (MEMO/08/533 of 31 July 2008). 17 Case COMP/39.402 RWE [2008] OJ C 133/10. 18 Cases COMP/39.388 German Electricity Wholesale Market, and COMP/39.389 German Electricity Balancing Market [2009] OJ C36/8 (E.ON cases). 19 20 Case COMP/39.315 ENI [2010] C 352/8. Case COMP/39.316 GDF [2010] C 57/13. 21 Case COMP/39.351 Swedish Interconnectors [2010] C 142/28. 12

10.3 A typology of antitrust issues related to capacity mechanisms

185

Commission went as far as classifying congestion shifting by the Swedish TSO as an abuse of a dominant position, and required the TSO to split the market into price zones, thereby changing the whole market design. By the third phase, the intensity of EU antitrust enforcement has lessened considerably due, in part, to the greater involvement of the national competition authorities (NCAs). The Commission and NCAs coordinate within the European Competition Network. The Commission now tends to focus on the most complex issues, such as power exchanges, and more political cases (see, for example, the on-going investigation against Gazprom).

10.3 A typology of antitrust issues related to capacity mechanisms This section aims to provide a framework for different antitrust issues that may occur in the context of capacity mechanisms. As described in detail in section 1.2.3, capacity mechanisms can be classified into five general categories: strategic reserve, capacity auctions, capacity obligations, reliability options, and capacity payments. Hybrid designs are also possible, increasing the range and complexity of possible antitrust issues. In particular, capacity obligations combine with secondary markets for capacity certificates (see chapter 14, discussing the French capacity mechanism). The initial concerns of capacity mechanisms relate to the behaviour of firms in auctions, and are discussed in depth in section 10.5.1, below. These issues are well known in energy markets where they have arisen, for instance, in the context of anticompetitive wholesale spot market bidding. In this case, withholding capacity or coordinated (collusive) bidding to increase prices is common. Bid rigging by dominant vertically-integrated incumbents may remain an issue in some cases. While our focus is on capacity auctions, where market abuse may have serious market-wide implications, we note that anticompetitive behaviour may also occur in auctions for the procurement of strategic reserve. A second set of issues arises when considering capacity mechanisms which involve bilateral contracting on a market parallel to the wholesale energy market. Bilateral contracting could take place in the context of both strategic reserve schemes without auctioning and capacity mechanisms with secondary certificate markets. In the first case, an independent agent, functioning as a sort of single buyer, enters into contract with generators for the provision of reserve capacity. The role of the individual agent is vital. As it is likely to be the TSO, its incentive scheme should be taken into account. In particular, an ineffective unbundled TSO could enter into contract with its affiliated supply or generation business on preferential terms, discriminating against other generators or suppliers. The length of contracts could also be an issue, if it is not regulated. If contracts are too short, new generation capacities will not enter the (capacity) market. The analysis provided by the Commission in its state aid decision on the UK capacity mechanism22 is notable in this regard. A useful point

22

UK capacity mechanism (n 4).

186

Antitrust Law: A Missing Piece in a Regulatory Puzzle? Table 10.1 Typology of antitrust issues in the context of capacity mechanisms. Capacity mechanism type

Possible antitrust issues (Articles 101 and 102 TFEU)

Strategic reserve

Gaming in auctions Anticompetitive bilateral contracting practices Anticompetitive practices by TSO Leveraging market power across markets

Capacity payments

Anticompetitive practices by TSO Withholding of capacities

Capacity auctions

Gaming in auctions Anticompetitive bilateral contracting practices Anticompetitive practices by TSO Leveraging market power across markets

Capacity obligations (with markets for certificates)

Market manipulation Margin squeeze Anticompetitive practices by TSO

Reliability options

Manipulation and insider trading in financial markets / under REMIT

Source: Authors’ own table.

of comparison is the different anticompetitive abuses discussed in the decisions concerning balancing markets.23 In the case of certificate markets, attempts to manipulate prices and increase costs for rivals, where the generator or supplier enjoys market power, would be a possibility (see chapter 14). More generally, the creation of a single trading platform is likely to be a superior solution to bilateral trading, as it would be more transparent and easier to monitor. The length of contracts could also be a concern, even though the problem of foreclosure does not seem to be as salient as it is in energy-only wholesale markets, as they would usually have shorter durations. A third set of issues revolves around the interaction between capacity mechanisms on the one hand, and wholesale and balancing markets on the other hand. It is likely that certain dominant market players would be able to leverage market power across markets. For instance, given that the strike price in a strategic reserve scheme constitutes an implicit price cap in the wholesale market, generators might have an incentive to manipulate the strategic reserve scheme to maximize profits on the wholesale market. One could also imagine dominant incumbents selling bundles of energy and capacity products, thereby creating anticompetitive tying issues when specific auctions are not organized. A possible solution is to prevent parallel participation in both wholesale markets and capacity mechanisms. A fourth set of issues, specific to reliability options, may involve the manipulation of financial markets and insider trading. It follows from the previous discussion that the range of anticompetitive practices raised by capacity mechanisms is limited, and that some types of designs might be more prone to anticompetitive conduct than others. Table 10.1 attempts to classify antitrust

23

E.ON cases (n 18).

10.4 The problem with the market definition

187

issues according to the design type. It is also clear that capacity mechanism designs often entail vesting an agent with particular responsibility, such as tendering for capacities. The TSO is likely to be the agent in question and its behaviour, as a consequence, needs to be suitably regulated. This will be discussed in more detail below, in section 10.5.2.

10.4 The problem with the market definition Before we focus on the selected anticompetitive practices, we would like to briefly discuss the issue of market definition. The assessment of competitive behaviour requires the identification of the (relevant) market in which such behaviour takes place. Broadly speaking, the definition of the relevant market sets the limits of rivalry or, in other words, the competitive pressure. The reason for defining the specific market and its relevance differs depending on whether it is done for the purpose of Articles 101 or 102 TFEU. In the former case, market definition helps to determine whether the agreement or concerted practice at issue constitutes an appreciable restriction of competition or cross-border trade, justifying an antitrust response, and is relevant to the application of ‘de minimis’ rule and group exemptions. However, hardcore cartels (such as price-fixing) constitute restriction to competition regardless of their appreciable effect, and the market definition is less relevant in such cases. To the contrary, market definition is of major importance in Article 102 cases. Defining the relevant market is essential in assessing whether a given behaviour comes from a dominant position, as dominance is always established in a given (relevant) market. Thus, it is far easier for the Commission to establish a dominant position, if it narrowly defines the market. The relevant market has two dimensions: a relevant product market, where products and services are substitutable, and a relevant geographic market, where competitive conditions are more or less homogeneous. The methodology of the Commission can be found in the Notice on the definition of the relevant market.24 In the electricity sector, the definition of the relevant market has always been a complicated issue, given the constant evolution of business models and the widening of geographical markets through increased regulatory convergence, and the physical capacity of interconnection. The relevant geographical market has generally been defined as national in scope in previous competition cases. It is not likely to change in the case of capacity mechanisms for at least two reasons. First, these schemes are defined more or less independently at national level. They tend to display strong differences in terms of design, and thus in terms of competitive conditions. Secondly, despite recommendations by the Commission, most capacity mechanisms will not be open to foreign companies, at least not in the short to medium term. The main reason is that, in view of physical and contractual congestions on interconnections, foreign generators are unlikely to be able to supply a reliable capacity product. Overcoming the congestion problem and ensuring a reasonable degree of regulatory convergence will take some time. We can therefore assume that the relevant geographical market will remain national in scope, for the time being. 24 Commission Notice on the definition of relevant market for the purposes of Community competition law [1997] OJ C 372/5.

188

Antitrust Law: A Missing Piece in a Regulatory Puzzle?

As regards the relevant product market, the main difficulty is to determine whether capacity mechanisms should be seen as a market distinct from generation and wholesale supply, ie the production of electricity as well as electricity physically imported through interconnectors, including a priori power exchanges. In certain earlier competition decisions, separate markets for wholesale trading and ancillary services have also been defined. For example, the E.ON case involves two decisions from the Commission with respect to two separate product markets, the wholesale market and balancing market.25 Wholesale supply to large end customers may also be seen as a separate product market, with these customers having very different patterns of consumptions from traders and resellers. There are indications in the decisional practice of the Commission that capacity mechanisms would be considered as a separate product market, for the following reasons. First, it is likely that the Commission will be sensitive to the fact that capacity mechanisms will probably be tightly regulated compared to other forms of wholesale supply. In the E.ON/MOL case26 for example, the Commission decided that the regulated part of the market, where generators had to sell to a single buyer for public utility purposes, was to be considered as distinct from a competitive wholesale supply. The mechanism, in that case, was similar to a strategic reserve scheme with an individual agent tendering or contracting for reserve capacities. It is interesting to note that no sub-national (ie regional) capacity mechanisms seem to have been anticipated. Secondly, the nature of the product sold in a capacity mechanism (capacity, MW) is different from the product sold in the wholesale market (electricity, MWh). Capacity products can even differ depending on the capacity mechanism. For example, the fact that strategic reserve schemes only concern peak hours increases product differentiation. The way capacity products are defined, in terms of blocks of capacity, the procurement method, and duration can also reinforce differentiation. Similarly certificates, in the context of a certificate market, are of a different nature than energy-only products, and other capacity products for that matter. In terms of reliability options, the question is whether they belong to a separate product market, different from wholesale trade in the non-physical financial delivery, or simply from the market for wholesale trading. Thirdly, the type of market players involved might also play a role. Given the specificities of capacity products, it is possible that certain agents, such as DSR providers, will only participate in a capacity mechanism, and will not be a part of the wholesale energy trade. Certain actors might also be excluded from capacity mechanisms for different reasons, for instance because a particular technology is, by definition, ill-suited to offer capacity products with certain specifications due to long lead-times. Fourthly, the trading venue may be a relevant concern. In the case of a hybrid design, for example, it could be argued that separate product markets exist within one capacity mechanism, for example between capacity products procured through bilateral contracts and more standardized capacity products procured through an organized platform. This will depend on whether buyers view them as substitutes, and thus, whether the two types of capacity products compete with each other. 25 26

See cases at n 18. Case COMP/M.3696 E.ON/MOL [2006] OJ L 253/20.

10.5 Selected anticompetitive practices under Arts 101 and 102 TFEU

189

Overall, it seems reasonably clear from the discussion in this section that a separate (national) specific market for capacity mechanisms is likely to be found. The fact that there will be a connection between wholesale consumption and capacity obligations, and that the position of the parties in the provision of capacity products could help strengthen their dominant positions in wholesale supply, could also be decisive factors. Overall, from a policy perspective, it is also tempting to argue that capacity mechanisms should be considered as a separate market for the purpose of applying Articles 101 and, in particular, 102 TFEU, in order to provide sufficient control over potential anticompetitive behaviour.

10.5 Selected anticompetitive practices under Articles 101 and 102 TFEU In this section we look, in more detail, at two selected anticompetitive practices that may be addressed under Articles 101 and 102 TFEU: the anticompetitive behaviour of firms in auctions (section 10.5.1) and the anticompetitive practices of TSOs (section 10.5.2). Our two examples aim to illustrate that antitrust enforcement is just one of several methods to tackle anticompetitive behaviour in the context of capacity mechanisms, and its effectiveness heavily depends on the type of market abuse. While in some cases antitrust actions constitute an adequate response, they might be less effective in other cases, and should be accompanied by other types of regulatory intervention, in particular regulation and market monitoring.

10.5.1 The anticompetitive behaviour of firms in capacity auctions There are valid theoretical reasons, supported by evidence from the US and Latin American capacity (and energy) auctions,27 for suspecting that capacity auctions may be susceptible to gaming. By gaming, we mean different types of strategic bidding to manipulate the outcome of the auction, for instance, by raising the price. Gaming in a capacity auction can fall under EU antitrust rules in two ways. It can be considered an abuse of a dominant position by a single generator, in the form of excessive bidding for example, and assessed under Article 102 TFEU. We illustrate this bidding strategy in section 10.5.1.1. Alternatively, it can be a result of collusive practice between several bidders, covered by Article 101 TFEU (see section 10.5.1.2).28 However, as shown below in section 10.5.1.3, antitrust law presents substantial limitations when it comes to prosecuting gaming in electricity wholesale markets, resulting in long, costly investigations, which are riddled with errors and lead to contestable decisions. Therefore, we

27 See chapter 7, and also David Harbord and Marco Pagnozzi, ‘Britain’s electricity capacity auctions: lessons from Colombia and New England’ (MPRA Paper No 56224, April 2014, mimeo). 28 Collusive practice can be also assessed under Article 102 TFEU as a special case of joint dominance. This approach has been used by the Commission in the E.ON case (n 18) with respect to the alleged gaming in the German electricity wholesale market. The E.ON case is discussed below in section 10.5.1.3.

190

Antitrust Law: A Missing Piece in a Regulatory Puzzle? Capacity price (EUR/MW) Pmax

1st round D 2nd round

3rd round

P net CONE S

Qmin

QT

Qmax

Capacity (MW)

Figure 10.1 Descending-clock, pay-as-clear capacity auction (schematic representation). Source: Authors’ own illustration.

expect that ex post antitrust investigations into gaming in capacity auctions will not be a frequent form of regulatory intervention. Nor should it be. Instead of tracing the actual abuse of market power, regulators should intervene ex ante, and look for the potential to exercise market power and ways to mitigate it. This involves both improving capacity auction design (section 10.5.1.4), and strengthening market monitoring (section 10.5.1.5). As discussed in chapter 7, there are many types of capacity auctions, and none of them is immune to gaming risks. However, it is beyond the scope of this chapter to analyse them one by one. Just for illustrative purposes, we refer to the UK capacity auction model (Figure 10.1), which is a periodic, descending-clock, pay-as-clear auction in which all successful generators are paid the last-accepted bid (S). The target level of capacity to be bought is set administratively (QT). The auctioneer announces a high price at the beginning of the auction (Pmax) and generators submit bids to indicate how much capacity they are willing to supply at that price. This process is repeated in successive rounds according to a pre-determined schedule until the auction discovers the lowest price at which demand (D) equals supply. All successful participants are paid the same clearing price (P), equal to an administrative estimate of the reasonable net cost of new entry (net CONE).29

29 Net CONE sets the price at which the target level of capacity would be auctioned. Net CONE reflects the cost of a new build CCGT (gross-CONE) minus expected electricity market revenue, and can be revised if necessary for each auction. The reason why P is set at the cost of a CCGT entry is because these are marginal plants. They run least, so they most need a capacity payment, and should therefore be setting the price in the auction. See chapter 22.

10.5 Selected anticompetitive practices under Arts 101 and 102 TFEU

191

10.5.1.1 Gaming as an abuse of a dominant position by a single generator In a pay-as-clear auction, where all winning generators are paid a uniform price for their capacity (pay-as-clear), large portfolio generators may have an incentive and the ability to unilaterally raise the market-clearing price. This strategy is known as capacity withholding, and comes in two forms: the physical withholding and economic withholding. In the former case, a generator literally opts out some of its capacity from the auction, either by not bidding, de-rating, or declaring unit outages. In the latter case, a generator simply submits an excessively high bid on some of its auctioned capacity, thereby excluding it from the auction. Whether physical or economic, increasing the auction price by withholding capacity is not a straightforward strategy. Not all generators would be able to trigger a price increase by withholding, and not all generators would have an incentive to do it. First of all, a generator is able to unilaterally affect the auction price by adjusting its bidding behaviour only when it is crucial to the continuation of the auction (a pivotal generator). What does this mean? Figure 10.1 illustrates that the margin of supply and demand decreases as the auction progresses. The auction can reach a point where one generator is left to arrive at the targeted capacity demand level. Its generation unit becomes strategic to the outcome of the auction (S). If this generator withholds capacity S, as presented in Figure 10.2, the auction closes early, at a market-clearing price equal to the last-accepted bid (PW). This price is higher than the market-clearing price without gaming (P equals net CONE). As a result of withholding by one generator, all generators successful at the auction (units to the left) receive higher capacity payments equal to PW. It is argued that a descending-clock auction with successive bidding rounds invites gaming, because generators can observe the changing balance of demand and supply after each round. This allows them to anticipate whether and when exactly they become pivotal (that is, critical) to the auction, and develop strategies for prematurely closing the auction at a higher price. Moreover, as capacity auctions are held on a yearly basis, there might be concerns that the repeated interaction of generators with similar supply and demand conditions would help them to ascertain their bidding patterns and exploit this information to behave strategically. Secondly, the profitability of capacity withholding is far from clear, and depends on a number of factors. Only large portfolio generators may find it profitable. Withholding implies a loss of revenue with respect to capacity removed from the auction in the form of forgone capacity payments. However, if a generator owns a substantial share of capacity which is successful at the auction (Figure 10.2, units to the left), the forgone profit from withheld capacity may be more than that which is offset by the rise in price for the remaining capacity. In sum, a large portfolio generator may have an incentive to strategically withhold capacity if the lost capacity payments on the withheld capacity are smaller than the additional revenue gained by the remaining plants in the auction. This will, in turn, depend on the quantity and the type of plant withheld compared to the plants remaining in the auction, and the slope of the demand curve (the steeper the slope, the greater the price increase). The incentive to engage in strategic withholding may be further strengthened by the opportunity to sell withheld capacity in a

192

Antitrust Law: A Missing Piece in a Regulatory Puzzle? Capacity price (EUR/MW) Pmax

1st round D 2nd round

3rd round P net CONE

Qmin

QT

Qmax

Capacity (MW)

Capacity price (EUR/MW) Pmax

1st round D 2nd round

PW

3rd round

P net CONE

Qmin

QW

QT

Qmax

Capacity (MW)

Figure 10.2 Capacity withholding in a descending-clock, pay-as-clear capacity auction (schematic representation). Source: Authors’ own illustration.

subsequent capacity auction or bilaterally. To the contrary, the high probability of detection multiplied by the expected penalty in a follow-up investigation would discourage gaming.

10.5.1.2 Gaming as a result of collusion The characteristics of the electricity markets, and the repetitive interaction of the generators, makes these markets particularly prone to collusion, the second type of

10.5 Selected anticompetitive practices under Arts 101 and 102 TFEU

193

gaming discussed in this chapter.30 Multi-round and periodic capacity auctions allow generators to understand the bidding patterns of their competitors, to coordinate their bidding strategies and, importantly, enable the punishment of deviations in future auctions or rounds. However, collusive gaming in auctions requires a certain degree of communication between generators, as their bids are not disclosed to other auction participants. Communication between generators can take place either ahead of the auction or during the auction. For instance, generators can make public or private statements about the intended bidding, thereby encouraging other generators to adjust their bidding behaviour on the basis of this information. If, for example, a generator withholds capacity to raise the auction price on the basis of information provided by another generator, this is collusion, even though the price increase is triggered by a unilateral action. Theoretically, joint bidding strategies are also possible, especially in an oligopolistic environment with few large bidders. The higher the number of bidders, the more difficult it becomes to implement a joint bidding strategy and enforce it.31

10.5.1.3 The limitations of antitrust enforcement The application of EU antitrust rules to gaming in capacity auctions presents specific problems. Let us first consider a unilateral gaming strategy to raise the auction price by withholding capacity. In order to capture this behaviour under Article 102 TFEU, the Commission first needs to prove that the generator has a dominant position (or market power in economic terms) in the relevant market, and then, that it has abused its dominant position (exercised market power) by engaging in price manipulation. As already explained in section 10.4, defining the relevant market is essential in the application of Article 102 TFEU, as it serves the Commission to establish whether an undertaking enjoys market power, which is usually done by a market share calculation and concentration indices in the relevant market.32 However, it is widely acknowledged that market shares are not a suitable measure of the market power in wholesale electricity markets.33 While it might be a good proxy to see if the generator is large enough to have an incentive for gaming (a large portfolio of plants participating in the auction), it ignores many other factors which can contribute to the exercise of market power, in particular, demand conditions, vertical integration, auction design and market contestability. More sophisticated techniques have thus been developed specifically to measure the market

30 OECD, Competition Issues in the Electricity Sector, Background Note, OECD Journal of Competition Law and Policy Volume 6–4 (2005) p 85; Bert Willems and Emmanuel De Corte, ‘Market power mitigation by regulating contract portfolio risk’ (2008) Energy Policy 36 (10), 3787–96. 31 Charles River Associates, Capacity market gaming and consistency assessment—final report, CRA Project No D18985-00, September 2013, prepared for DECC (‘CRA Report’), pp 26–8. Market Analysis Ltd, Market Power in Electricity Markets: Do Electricity Markets Require Special Regulatory Rules? (2000) pp 25–7. Natalia Fabra, ‘Uniform pricing facilitates collusion: the case of electricity markets’ (mimeo, European University Institute, 1999). 32 A commonly accepted measure of market concentration is the Herfindahl-Hirschman Index (HHI), which is calculated by squaring the market share of each firm competing in the relevant market, and then adding together the resulting numbers. 33 David Newbery et al, ‘A review of the monitoring of market power’ (ETSO Report, November 2004) p 5.

194

Antitrust Law: A Missing Piece in a Regulatory Puzzle?

power in electricity markets.34 While competition authorities can employ a range of electricity-specific indices in antitrust investigations into gaming, it is less certain whether they would be accepted in court as a proof of dominance. The temporary nature of the market power in capacity auctions may not fit the court’s assessment of dominance, especially if a crude market share calculation does not point to a dominant position. Finding an abuse of a dominant position, which is the second step of an antitrust investigation, is no less problematic. First of all, it is difficult to assess whether the allegedly manipulated auction price is an excessive price, as it is difficult to determine costs and establish the appropriate competitive ‘reference’ levels. Just to give an example, in the UK capacity auction model,35 the competitive price is set at the level of net CONE. Net CONE is simply an estimate of the reasonable cost of new entry, based on the cost of a new build CCGT plant minus the expected electricity market revenue, and can be revised if necessary for each auction. It has been established that the existing plants can bid only up to 50 per cent of net CONE, and higher bids may be subject to investigation by OFGEM. In such cases, the generator would need to justify that the higher bid results from its net going-forward costs.36 However, not all bids which deviate from the estimated net going-forward costs necessarily imply gaming, but can simply result from the process of price discovery in a multi-round auction. As the auction progresses, generators learn about the energy market revenue expectations of other bidders. This affects their valuation and is reflected in their bids.37 Equally problematic is finding evidence of withholding, that is, identifying capacity which could have been sold profitably in the auction, but has been withheld from the sale. There are methods to identify this ‘gap’ in output, as illustrated in Figure 10.2, but they all require the actual auditing of de-rating and outages, and taking account of many different factors included in production decisions (such as ramp rates). This means handling a great amount of data in order to come up with results which are open to misinterpretation. As explained earlier, gaming can be a result of collusion. Article 101 TFEU prohibits not only anticompetitive agreements between firms, but also less explicit forms of concerted practices. For the purpose of Article 101 TFEU, the Commission does not need to prove a dominant position, however this does not mean that they will have to adopt a less stringent standard of proof. If the Commission develops a theory of collusive gaming, it needs to show evidence of an agreement or concerted practice. This, however, can be extremely difficult. Sophisticated game-theoretic models of oligopoly interaction can support the case, but may not replace the evidence of communication between the generators. The major drawback of relying on these models is that they come with a number of assumptions, questioning the relevance 34 In particular, the Residual Supply Index (RSI), Pivotal Supplier Index (PSI), and residual demand analysis. These methods are expected to become a standard technique in the market power analysis of electricity markets. See David Newbery et al (n 33) p 41. 35 See chapter 22. 36 A net going-forward cost is the cost a generator incurs in order to run a power plant in a year with a net capacity obligation of anticipated market earning. Linking bids to the net going-forward cost means that the plants do not seek a higher price than the one allowing them to stay operational in order to deliver their capacity commitment. 37 See CRA Report (n 31) para 103.

10.5 Selected anticompetitive practices under Arts 101 and 102 TFEU

195

of quantitative conclusions.38 Of course, antitrust interventions remain important in clear-cut cartel cases. Where communication between generators is explicit and can be proven, for example, by pointing to an agreement or other ‘smoking gun’ evidence (recorded phone calls, public statements, minutes from the meetings), the antitrust response is adequate and desirable. In sum, finding evidence of gaming behaviour in a capacity auction and proving an infringement of competition rules is particularly challenging. One could question whether the Commission or national competition authorities are well-equipped to carry out such complex and time-consuming antitrust investigations. The 2008 E.ON case, where the Commission investigated the German electricity spot market in search of capacity withholding, is a telling one. The case was closed by a commitment decision, so the Commission was not required to actually find an abuse of a dominant position. The Commission’s theory of harm was based on physical capacity withholding by E.ON to increase prices at the German power exchange (EEX) as a result of collusion with RWE, EnBW and possibly Vattenfall.39 The evidence for capacity withholding was based on the calculation of load factors for German power plants, carried out during the energy sector inquiry. The results from the inquiry pointed to noticeable mark-ups, over and above the competitive price benchmark, and significant discrepancies between the load factors of plants with similar marginal costs, just below the market clearing-price level, implying that some plants did not run at full capacity.40 It is clear that while gaming could be one possible explanation behind these results, there are other plausible reasons for the lower capacity factor of certain units, such as trading in other markets. A year after the Commission’s decision in the E.ON case, the German Federal Cartel Office (FCO) carried out its own investigation into gaming at EEX in 2007–2008. The FCO examined the bidding behaviour of four generators suspected of collusion in the E.ON case. A special algorithm was developed in order to detect the patterns of withholding in the generators’ day-ahead bids. Finally, in 2011, the FCO published a comprehensive report from its inquiry, but found no evidence of abusive practices.41 Even though the analysis of day-ahead bids suggested that a small share of capacity was not put in operation at times when it was profitable, the FCO admitted that there are other possible explanations behind the output ‘gap’. It concluded that: Abusive practices of this kind are extremely difficult to prove. To do so would require extensive data on the operations of each of the 340 electricity generating units over lengthy periods as well as the opportunity to more effectively check these company data [sic] and information on marginal costs, which are subject to frequent variation and include a large number of individual cost items. Also, the fact that the undertakings do 38 For an overview of the techniques available to detect the market power in electricity markets, see Newbery et al (n 33), Table 3 at p 46. 39 The Commission argued joint dominance. See n 28. 40 DG COMP Report (n 14) paras 428–448. See also London Economics and Global Energy Decisions, Structure and performance of six wholesale electricity markets in 2003, 2004 and 2005, Study presented to DG Competition on 26 February 2007, p 17 of Executive Summary. 41 FCO, Sector inquiry into electricity generation and wholesale markets (Report (B10-9/09, January 2011).

196

Antitrust Law: A Missing Piece in a Regulatory Puzzle?

not offer power plant capacity individually from each particular generating unit, but on the basis of complex stochastic optimisation from a pool of generating units also poses particular challenges.42

The FCO emphasized the importance of continuous market monitoring in order to prevent withholding strategies being deployed. For this purpose, it created a market transparency office within its structure in 2012, which now continuously monitors price formation in the German electricity wholesale market.43 The case of Germany discussed above demonstrates a welcome change in the regulator’s approach to market abuse in electricity markets. Evidence of price manipulation strategies by generators is rarely conclusive and often exposed to errors in interpretation, which makes antitrust law an ill-suited framework for detecting gaming in electricity or capacity auctions, in most cases. We can thus observe a welcome shift towards market surveillance. We believe ex ante forms of regulatory intervention can better address gaming in capacity auctions than ex post antitrust enforcement. A careful auction design should be accompanied by strengthened market monitoring.

10.5.1.4 Ex ante intervention through auction design Gaming would not be profitable in a truly competitive capacity auction—raising the bidding price or physically withholding output would just result in a smaller market share, without receiving any additional revenue on the rest of the generator’s portfolio. If the market in Figure 10.1 was contestable, withholding capacity S would not trigger any price increase. Even though the existing capacity is fixed, withheld capacity would be replaced by the bid of a new entrant at price P, which equals net CONE. Therefore, a point of departure for any auction design would be to increase contestability as much as possible. An auction design should not unnecessarily restrict entry or discriminate against smaller players. The most obvious entry barrier is prequalification. This should not go beyond the minimum necessary to establish a group of capacity providers technically able to perform at times of system stress. This means opening the auction to different technologies, both existing capacity and new investments. In practice, the objective of creating competition often clashes with other policy objectives pursued by the regulator. For instance, the regulator may choose to exclude certain types of technologies from the auction, either because, for example, they are polluting (coal-fired generation) or receive support from another scheme (RES, nuclear). Such policy choices, justifiable for other reasons,44 would inevitably restrict competition in the auction. Apart from pre-qualification, there are less obvious design elements which are nevertheless fundamental from a competition perspective, as they may de facto hamper the entry of new investments, certain types of technologies, or smaller players. At the 42

FCO (n 41) p 19. The FCO is integrated in the new EU market surveillance scheme created under REMIT, and closely cooperates with the Federal Network Agency (BNetzA). 44 Climate policy objectives would justify excluding polluting plants, whereas already subsidized RES may be excluded from capacity auctions to avoid any risk of double payments. 43

10.5 Selected anticompetitive practices under Arts 101 and 102 TFEU

197

outset, to encourage the participation of new investments, the auction should provide sufficient time between the auction and the delivery year, allowing for construction (lead time), and grant longer contracts for new investment to enable cost recovery. For example, the UK capacity auction has a lead time of four years and offers longer contracts for new generation (fifteen years). The competitive price benchmark is based on the cost of the CCGT technology. A combination of these features is supposed to ensure that new CCGT plants enter the auction.45 What about other technologies, which would need longer or shorter lead times? Shouldn’t they have an opportunity to submit bids with different lead times? Ensuring that there is no de facto discrimination between technologies in a single auction adds to its complexity and diminishes transparency: how are bids with different lead times to be compared? As discussed in section 7.4 above, separate, technology-specific auctions may be a simpler and more transparent solution, often reflecting the energy policy choice, in order to foster the growth of a particular technology.46 However, this inevitably limits competition in a capacity auction to providers of one particular technology, and, from a purely economic perspective, may lead to a suboptimal generation mix in the long run.47 The UK capacity mechanisms allow for separate one year auctions for DSR providers, which would find it difficult to make demand-side commitments four years in advance. Lastly, the same contract terms for larger and smaller capacity providers may put the latter group in a disadvantaged position. For instance, generators with a range of power plants are better equipped to handle the risks associated with non-performance compared to smaller generators which, due to a limited asset portfolio, cannot easily cover for the loss of a plant in a stress event. The greater risk factor for small generators would thus be reflected in their (higher) bids. A capacity mechanism design can address this problem by, for example, allowing different forms of secondary trading between the auction and delivery, so that generators can adjust their positions to better manage their exposure to penalties for non-delivery. In that respect, the UK capacity mechanism envisages secondary financial trading, volume reallocation and obligation trading.48 There are also monthly and annual caps on penalties, which are linked to the provider’s capacity revenue, and therefore their size. One step further would be to develop a penalty insurance market, so that generators can simply insure themselves against the risk of non-performance.49 Certain design features of the UK capacity mechanism have been introduced specifically to mitigate gaming risks. Most importantly, the auction includes a cap and a sloped demand curve, so that less capacity is procured in a given year if the price 45 See chapter 22. However, the first capacity auction held in December 2014 was cleared at a price of 19 £/kW, much below the estimated cost of entry for a new CCGT power plant (49 £/kW). 46 For instance, separate year-ahead auctions for DSR providers. 47 However, a given generation mix might be politically desirable, reflecting other policy objectives (climate policy, developing technologies based on domestic resources, employment policy, among others). 48 See chapter 22. For details, see Capacity Market Rules 2014, as amended by the Capacity Market (Amendment No 1) Rules 2014 and the Capacity Market (Amendment No 2) Rules 2014, available at https://www.gov.uk/government/publications/capacity-market-rules, accessed 1 February 2015, chs 9 and 10. 49 CRA Report (n 31) para 16.

198

Antitrust Law: A Missing Piece in a Regulatory Puzzle?

is very high. This reduces the incentives to withhold capacity, as price increases are small in relation to the output reductions which trigger them. Furthermore, existing plants can bid only up to a certain threshold (50 per cent of net CONE). If an existing plant bids above this threshold, it may be subject to investigation by OFGEM, and would need to justify higher capacity payments by its net going-forward costs. It is anticipated that the level of supply in each auction and the identity of particular bidders will be concealed, to mitigate the risk of collusion.50 Lastly, there will be a monitoring and compliance mechanism in place, which shall be discussed next.

10.5.1.5 Ex ante intervention through market monitoring and enforcing REMIT Auction rules can alleviate the risk of market abuses to some extent, but cannot protect against all potential gaming risks. Market monitoring can fill this gap, and there is a growing consensus that it has become essential to a well-functioning electricity market. A more systematic market monitoring framework at EU level has only recently been introduced. REMIT came into force in December 2011, and stipulates an explicit prohibition of market manipulation, attempted market manipulation and insider trading, together with alternative (to antitrust) enforcement tools. Member States were required under REMIT to endow energy regulators with sufficient investigatory and prosecutorial powers to act upon these prohibitions.51 The UK auction rules include a ban on market manipulation and impose information barriers to limit insider trading.52 The definitions of these terms are those used in REMIT. Participants in the capacity auction are required to sign a Certificate of Ethical Conduct confirming that, among other conditions, they have not engaged in market manipulation. A third-party auction monitor will be appointed to verify that the rules are followed, and will report any potential breaches or suspected gaming behaviour, which can lead to formal investigations with respect to individual auction participants. As envisaged in the UK model, it is expected that the prosecution of gaming in the context of national capacity mechanisms will be carried out within the national legislative frameworks created to enforce REMIT, rather than by the Commission under the EU antitrust rules. There are, at least, two reasons in favour of this approach. First, the prohibition of market manipulation under REMIT does not require the national regulatory authorities to ascertain the dominant position, and more easily allows the peculiarities of the electricity markets to be assessed. Secondly, national energy regulators have sector-specific expertise and are better informed about their national markets than the Commission. They also have immediate access to the relevant market data and can cooperate with national competition authorities in enforcing REMIT.53 As such, they may be better equipped to prosecute market abuses 50

See chapter 22 for other measures to mitigate the scope for gaming. 52 REMIT (n 5) Art 13. See chapter 22. 53 REMIT envisages different forms of cooperation between energy regulators and national competition authorities. In particular, they can set up joint market monitoring and other forms of cooperation in carrying out market investigations and enforcing REMIT. Further, energy regulators are obliged under REMIT to inform competition authorities, in case they suspect breaches of competition rules on wholesale energy markets. See REMIT (n 5) Art 7, Art 16(1), Art 16(3)(d). 51

10.5 Selected anticompetitive practices under Arts 101 and 102 TFEU

199

in national capacity mechanisms. Another question is whether the Commission is willing to step back from investigating gaming behaviour in electricity and capacity auctions, leaving the market in the hands of national regulatory authorities. It must be noted that REMIT does not preclude the Commission’s antitrust intervention, should a national authority fail to prosecute market abuses in capacity auctions.

10.5.2 The anticompetitive practices of TSOs TSOs are expected to be the central players in most capacity mechanism designs. They are likely to be the agent procuring the capacity products in strategic reserve and capacity auction schemes. They might be involved in the pre-qualification and certification of operators participating in capacity mechanisms, as well as the determination of availability commitments. They are likely to be consulted, if not vested with a decision-making role, on the definition of the capacity product itself, or the capacity obligation per supplier. They are also likely to hold advisory roles to the regulator on a range of topics, such as administrative price setting. They could even be vested with non-monetary sanction powers. There is thus a wealth of opportunity for a non-benevolent TSO to distort the competitive process. It seems unlikely that a TSO could favour its affiliated generation arm in a tendering process without difficulty, which is another argument for the use of auctions over bilateral contracting in the context of capacity mechanisms. However, the inherent asymmetry of information and even competences between the regulator and the TSO54 may facilitate subtle exclusionary practices, particularly in the way capacity products or obligations are specified, or the way pre-qualification, certification, and financial settlements are conducted. Another issue concerns the management of interconnections, in case access to the capacity mechanisms is allowed for foreign generators. The TSOs indeed have the means to trick the cross-border transmission capacity allocation process,55 or artificially increase cross-border physical congestions.56 If interconnectors co-owned by a national TSO are allowed to provide capacity products themselves, that very same TSO might be procuring and supplying capacity, thereby creating potential conflicts of interest. The TSOs, especially (but not only) ineffective unbundled TSOs, have a long record of (mainly network-related) anticompetitive abuses in the energy sector. Their ability to introduce distortive biases in capacity mechanisms should not be underestimated. The question of a suitable regulatory framework for TSOs in the context of capacity mechanism implementation is therefore an issue that deserves the careful consideration of regulators. 54 See for example, Adrien De Hauteclocque and Vincent Rious, ‘Reconsidering the regulation of merchant transmission investment in light of the third energy package: the role of dominant generators’ (2011) Energy Policy 39 (11), 7068–77. 55 See on this, Hauteclocque (n 13). This might be relevant even if it is expected to allocate priority access rights to foreign participants. 56 See two articles on the Swedish Interconnectors case by Bert Willems and Małgorzata Sadowska, ‘Power markets shaped by antitrust’ (2013) European Competition Journal 9 (1); ‘Market integration and economic efficiency at conflict? Commitments in the Swedish interconnectors case’ (2013) World Competition: Law and Economics Review 36 (1).

200

Antitrust Law: A Missing Piece in a Regulatory Puzzle?

The enforcement of antitrust rules as a regulatory instrument can be instrumental, especially if it is not used in isolation. Despite asymmetries of information and high sector specificity, antitrust authorities have a considerably strong history of tackling TSO abuses, both at Union and national levels. There are obvious benefits in searching for better synchronization between sector regulation and antitrust enforcement in this case. The sector regulator remains better placed to analyse certain technical characteristics of the industry, like the design of a capacity product or the determination of availability commitments. At the same time, an antitrust policy has a strong deterrence potential and can always be used as a regulatory patch after the first intervention of a sector regulator. Having a more integrated approach to competition policy, by harnessing the shared interests between sector regulation and antitrust powers, appears to be a suitable way forward.

10.6 Conclusions This chapter is a first attempt at systematizing potential antitrust abuses in the context of capacity mechanisms. We classify them according to capacity mechanism types, and focus on two selected anticompetitive practices, as apparent in capacity auctions and in relation to the TSOs, to discuss the effectiveness of antitrust rules in tackling them. While much is yet to be resolved, the continuing implementation of capacity mechanisms at national level will bring greater clarity to the issue. However, certain conclusions can already be drawn. It is clear that certain capacity mechanism types are more prone to anticompetitive behaviour than others. While capacity auctions seem generally superior to bilateral contracting, they are also not immune to gaming risks. The devil is in the (design) detail, and the choice of a specific capacity mechanism type should take into account the pre-existing market structure and the regulatory context, that is, enforcement powers and the relative strengths and weaknesses of national energy regulators, as well as competition and financial authorities. In particular, introducing a capacity mechanism in highly concentrated markets, with vertical integration or ineffective unbundling, should be accompanied by a careful impact assessment as to their likely anticompetitive effects on the market and the availability of national resources to manage them. As illustrated in this chapter, antitrust enforcement has its limits when it comes to tackling abusive behaviour in electricity markets. We believe that it can be effective, if accompanied by strong ex ante regulation, in particular, a careful capacity mechanism design and continuous monitoring.

11 Free Movement of Goods in the Labyrinth of Energy Policy and Capacity Mechanisms Peter Oliver1

11.1 Introduction Under the Treaty of Rome, the common market, and more particularly the customs union, was the cornerstone of the European Economic Community (EEC), which began life in 1958. As testimony to its importance, the draftsmen of that Treaty placed the provisions on the customs union, including those on the free movement of goods between Member States, immediately after the eight introductory articles. Today, the Treaty of Rome has been replaced by the Treaty on European Union (TEU) and TFEU2 and the EEC has been transformed into the European Union. Nevertheless, the common market (now known as the internal or single market), which is comprised of the free movement of goods, persons, services and capital as well as competition, still lies at the very heart of the EU. Without the single market, the EU could simply not exist. Reading some of the recent publications relating to the common energy policy, one gets a sense that, for lawyers and policy makers in this field, the free movement of goods under the EU Treaties is a long lost treasure—to my mind, it is a treasure, despite the endless debates about some issues—like Angkor Wat or Machu Picchu waiting to be ‘rediscovered’ by a latter-day Henri Mouhot or Hiram Bingham III. Far be it from me to attempt to cast myself in that role! In any case, I am no position to ‘rediscover’ this treasure since, like most EU lawyers, I never lost sight of it in the first place. Rather, my task is to elucidate the decided cases and indicate how they might influence the current debate on capacity mechanisms. Before turning to the case law, it is as well to recall that, according to Article 194(1)(b) TFEU, among the aims of EU energy policy is ‘in a spirit of solidarity between Member States, to . . . (b) ensure security of energy supply in the Union . . . ’.3 Have the words italicized here occasionally been overlooked? Recently, several large heaped spoonfuls of spice have been added to the discussion of this issue by AG Bot’s Opinion in Essent Belgium4 and more particularly by his more 1 The author wishes to thank Katarzyna Herrmann and Angus Johnston for their very helpful comments on an earlier draft of this chapter. However, the views expressed here are my personal views and any errors are attributable to me alone. 2 Treaty on the Functioning of the European Union [2012] OJ C 326/47 (hereinafter ‘TFEU’ or ‘the Treaty’). 3 Emphasis added. 4 Opinion of AG Bot in Joined Cases C-204/12 to C-208/12 Essent Belgium NV v Vlaamse Reguleringsinstantie voor de Elektriciteits—en Gasmarkt (judgment of 11 September 2014, nyr).

202

Free Movement of Goods

forthright Opinion in Ålands Vindkraft.5 In both cases, as we shall see, the Court has failed to heed him. This is all the more regrettable in that, as is plain from ACER’s Report6 and chapter 2, barriers to trade between Member States appear just as problematic as regards capacity mechanisms as in every other economic sector— which would have certainly not have surprised the draftsmen of the Treaty of Rome!

11.2 Energy as goods As is well known, in Commission v Italy the Court defined ‘goods’ for the purposes of the provisions of the Treaty of Rome relating to the free movement of goods as ‘products which can be valued in money and which are capable as such of forming the subject of commercial transactions.’7 On the basis of this definition, the Court famously dismissed Italy’s claim that works of art were too exalted to fall within the concept of ‘goods’; and it therefore held that an export tax on works of art fell foul of what is now Article 30 TFEU which prohibits customs duties on exports and charges of equivalent effect. There was never any doubt that all forms of energy other than electricity fall squarely within this definition. Electricity, on the other hand, lacks some of the qualities normally associated with goods: it is not tangible (in the normal sense!); and it is hard to store. Nevertheless, in the celebrated case of Costa v ENEL8 the Court, following the Advocate General, found that an electricity monopoly fell under what is now Article 37 TFEU. Since that provision relates exclusively to monopolies in goods, this ruling necessarily implied that electricity is to be regarded as goods, although neither the Court nor the Advocate General addressed this issue. In Almelo,9 the Court confirmed that electricity constitutes goods for these purposes, giving three reasons to support this conclusion: that this view was accepted both in EU law and in the laws of the Member States; that electricity was treated as a product in the customs nomenclature; and that it had already ruled to this effect in Costa v ENEL.10 In a subsequent case, AG Fennelly acknowledged that it might ‘appear somewhat surprising that the Court has treated electricity, despite its intangible character, as goods’.11 After repeating the reasoning of the Court in Almelo, he added: ‘To my mind, electricity must be regarded as a specific case, perhaps justifiable by virtue of its function as an energy source and, therefore, in competition with gas and oil.’12

5

Case C-573/12 Ålands Vindkraft AB v Energimyndigheten (judgment of 1 July 2014); Opinion of AG Bot of 28 January 2014. 6 ACER, Report on capacity remuneration mechanisms and the internal market for electricity, 30 July 2013 (ACER’s Report). For a comprehensive discussion on ACER’s Report, see chapter 2. 7 Case 7/68 Commission v Italy [1968] ECR 423, 428. 8 Case 6/64 Costa v Enel [1964] ECR 585. 9 Case C-393/92 Almelo v Energiebedrijf Ijsselmij [1994] ECR I-1477. 10 See also Cases C-158/94 Commission v Italy (electricity monopoly) [1997] ECR I-5789, paras 14–20, and C-206/06 Essent Netwerk Noord v Aluminium Delfzijl [2008] ECR I-5497, para 43. 11 Case C-97/98 Jägerskiöld v Gustafsson [1999] ECR I-7319, 7328–9. 12 Jägerskiöld v Gustafsson (n 11).

11.4 Free movement of goods: Restrictions

203

The latter reason appears particularly convincing. In any case, it is clearly established that restrictions on the movement of all forms of energy are caught by the Treaty provisions on goods (Articles 28 to 37 TFEU), not those relating to services.

11.3 Free movement of goods and state aid At this stage, it is helpful to consider very briefly the interaction between the various Treaty provisions on goods and those on state aids (now Articles 107 and 108 TFEU). It follows from the time-honoured ruling in Iannelli v Meroni that the two sets of provisions do not apply simultaneously, but that Article 34 TFEU does apply to a condition attached to the grant of an aid which is ‘not clearly necessary for attainment of its object or for its proper functioning’.13 By the same token, a severable condition attached to a state aid may fall foul of Articles 30, 35, or 110 TFEU.

11.4 Free movement of goods: Restrictions 11.4.1 General In this section, we shall consider in turn the Treaty provisions prohibiting restrictions on the free movement of goods between Member States, beginning with those concerning customs duties, charges and taxes (Articles 30 and 110 TFEU) and then moving on to non-tariff barriers to imports and exports (Articles 34 and 35 TFEU respectively). Even if a measure falls under Articles 34 or 35, it is compatible with the Treaties if it is justified on a ground recognized in EU law and meets various other conditions. Justification will be considered separately in section 11.5 below.

11.4.2 Article 30 TFEU Article 30 TFEU prohibits customs duties and charges of equivalent effect on imports and exports. In its seminal judgment in Social Fonds voor de Diamantarbeiders v Brachfeld, the Court held: ‘any pecuniary charge, however small and whatever its designation and mode of application, which is imposed unilaterally on domestic or foreign goods by reason of the fact that they cross a frontier, and which is not a customs duty in the strict sense, constitutes a charge having equivalent effect within the meaning of [Article 30] of the Treaty, even if it is not imposed for the benefit of the State, is not discriminatory or protective in effect or if the product on which the charge is imposed is not in competition with any domestic product.’14 The better view is that ‘a customs duty in the strict sense’ is a tax on imports and/or exports which is officially 13

Case 74/76 Iannelli & Volpi v Meroni [1977] ECR 557, para 9 ff. The citation is taken from para 14. Cases 2 and 3/69 Social Fonds voor de Diamantarbeiders v Brachfeld [1969] ECR 211, paras 15–18. See also Essent Network Noord (n 10) para 41. See Catherine Barnard, The Substantive Law of the EU—the Four Freedoms, 4th edn (Oxford: Oxford University Press, 2013) p 47 ff, Paul Craig and Gráinne de Búrca, EU Law—Text, Cases and Materials, 5th edn (Oxford: Oxford University Press, 2011), p 612 ff, and Peter Oliver and Martín Martínez Navarro, ‘Free Movement of goods’ in Catherine Barnard and Steve Peers (eds), European Union Law (Oxford: Oxford University Press, 2014) p 329 ff. 14

204

Free Movement of Goods

designated as a customs duty. Since they are prohibited, it would be most surprising if we were to encounter many of them today. Annex C to ACER’s Report15 states that ‘an example of a short term distortive effect of a [capacity mechanism] can be illustrated on the Finnish-Russian border’. According to this account, Russia’s capacity mechanism imposes a fee on exports of electricity to Finland during peak hours. To judge from the Report, it would seem that this fee applies by reason of the fact that the electricity ‘crosses a frontier’. Indeed, the fee would appear to bear all the hallmarks of a charge of equivalent effect except one: Russia is not a Member State. If it were, this fee would almost certainly fall foul of Article 30 TFEU. What is more, a breach of Article 30 TFEU will occur where the same charge is levied on domestic and imported electricity but the entire proceeds of that charge are paid to the producers of domestic electricity.16 If a tax is caught by Article 30 TFEU, the chances of saving it are practically nonexistent. Article 36 TFEU does not apply, as is plain from its wording.17 Admittedly, the Court has recognized two situations in which such charges are permissible, namely: (a) where the payment is consideration for a service rendered or (b) where it relates to inspections required by EU law, as where veterinary inspections are required by an EU Directive.18 However, the scope of these ‘exceptions’ is extremely limited and neither is likely to be of any avail to Member States which impose customs duties or charges of equivalent effect on imports and exports of energy. As to (a), this only arises where the individual importer or exporter receives a specific advantage by reason of the service; it does not suffice where the service is simply in the public interest.19 Moreover, the charge must be proportionate to the benefit conferred.20 As to (b), this has been held to apply only where the following conditions are fulfilled: (b1) the charge must not exceed the actual cost of the inspection, (b2) the inspections in question must be obligatory and uniform for all products in the EU, (b3) the inspections must be required by EU law, and (b4) they must promote the free movement of goods by eliminating obstacles which could arise from unilateral measures of inspection adopted by Member States in accordance with Article 36 TFEU.21

11.4.3 Article 110 TFEU In contrast to Article 30 TFEU, Article 110 TFEU prohibits discriminatory internal taxation: the former applies to charges imposed due to the fact that products cross a frontier, whereas the latter relates to taxes imposed within a Member State.22 It is settled law that a tax cannot fall under both provisions simultaneously.23 15

16 ACER’s Report (n 6). Essent Network Noord (n 10) para 57. Commission v Italy (n 7). 18 Case 46/76 Bauhuis [1977] ECR 5. The inspections themselves fall under Article 34 TFEU. 19 Case 24/68 Commission v Italy (statistical levies) [1969] ECR 193. 20 Case 132/82 Commission v Belgium [1983] ECR 1649, para 8. 21 Case 18/87 Commission v Germany [1988] ECR 5427, para 8. 22 Barnard and Peers (n 14) p 51 ff, Craig and de Búrca (n 14) p 621 ff. 23 Cases C-90/94 Haahr Petroleum [1997] ECR I-4085, para 19, C-28/96 Fazenda Pública v Fricarnes [1997] ECR I-4939, para 19, and C-383/01 De Danske Bilimportorer [2003] ECR I-6065. 17

11.4 Free movement of goods: Restrictions

205

Article 110 has been placed at quite some distance from Part III, Title II of the TFEU, which covers the free movement of goods (Articles 28 to 37 TFEU). Nevertheless, it relates exclusively to ‘products’, a term which in the English version of the Treaties, is used interchangeably with ‘goods’.24 What is more, the Court has gradually aligned Article 110 TFEU with the rules governing the free movement of goods. This has been effected in particular by two most welcome instances of judicial activism. First, although this is nowhere spelt out in the Treaty, the Court has held that this provision extends to goods originating in third countries but in free circulation in the Member States.25 Second, it has ruled that discrimination against exports in favour of products intended for domestic consumption is also contrary to this article.26 Accordingly, Article 110 TFEU has de facto become a provision relating to the free movement of goods. The locus classicus on Article 110 TFEU and schemes to boost the use of renewable energy is Outokumpu.27 That case concerned Finnish excise duties on electricity. Those imposed on electricity produced within Finland were calculated according to the method of production. In its reference for a preliminary ruling, the national court stated that the purpose of this scheme was to protect the environment. Thus electricity produced by water power was subject to a lower rate than that deriving from other sources. In contrast, imported electricity was subject to a flat rate. The excise duty chargeable on imported electricity was higher than the lowest rate due on electricity produced in Finland, but lower than the highest rate due on electricity within that Member State. The Court pointed out that it had ‘consistently held that that provision is infringed where the taxation on the imported product and that on the similar domestic product are calculated in a different manner on the basis of different criteria which lead, if only in certain cases, to higher taxation being imposed on the imported product (see, in particular, Case C-152/89 Commission v Luxembourg [1991] ECR I-3141, paragraph 20)’.28 At the same time, in accordance with its earlier case law, the Court held that it was compatible with Article 110 TFEU for a Member State to impose tax at rates which vary according to the method of production or the raw materials used if the scheme ‘pursues objectives which are themselves compatible with the requirements of the Treaty and its secondary legislation, and if the detailed rules are such as to avoid any form of discrimination, direct or indirect, against imports from other Member States or any form of protection of competing domestic products’.29 Moreover, the Court went on to rule that the protection of the environment was a legitimate basis for imposing differentiated tax rates. On this point, it referred inter alia to its earlier judgment in 24 Whereas in certain cases such as the heading of Part III, Title II and Art 28(1), the TFEU uses the term ‘goods’, in others such as Arts 28(2) TFEU and 29 the expression ‘products’ has been preferred. Similarly, the French text uses ‘marchandises’ and ‘produits’, the Italian ‘merci ’ and ‘prodotti’ and the Dutch ‘goederen’ and ‘produkten’. On the other hand, the German version uses ‘Waren’ throughout. 25 Case 193/85 Co-Frutta [1987] ECR 2085. 26 Case 142/77 Larsen [1978] ECR 1543, paras 24–26. 27 Case C-213/96 Outokumpu Oy [1998] ECR I-1777. 28 Outokumpu (n 27) para 34. Emphasis added. 29 Outokumpu (n 27) paras 30–1. For a graphic example of differential taxation which was discriminatory, see Case 112/84 Humblot [1985] ECR 1367.

206

Free Movement of Goods

Commission v Denmark (returnable bottles) where it had held that measures falling under what is now Article 34 TFEU can be justified on the grounds of environmental protection.30 Both AG Jacobs and the Court accepted Finland’s claim that it was impossible to determine the origin of electricity once it entered the distribution network and that it was therefore impossible to introduce the same system of differentiated rates for domestic and imported products. However, while the Advocate General maintained that the scheme should be upheld on environmental grounds,31 the Court took the opposite view, stating: ‘The Court has already had occasion to point out that practical difficulties cannot justify the application of internal taxation which discriminates against products from other Member States (see, inter alia, Case C-375/95 Commission v Greece [1997] ECR I-5981, paragraph 47).’ While the characteristics of electricity may indeed make it extremely difficult to determine precisely the method of production of imported electricity and hence the primary energy sources used for that purpose, the Finnish legislation at issue did not even give the importer the opportunity of demonstrating that the electricity imported by him had been produced by a particular method in order to qualify for the rate applicable to electricity of domestic origin produced by the same method.32 Finally, as we have already noticed, where the same charge is levied on domestic and imported electricity but the entire proceeds of that charge are paid to the producers of domestic electricity, that constitutes a breach of Article 30 TFEU.33 On the other hand, where only part of the proceeds are paid to the domestic electricity producers, Article 110 TFEU is breached.34

11.4.4 Article 34 TFEU Article 34 TFEU reads: ‘Quantitative restrictions and all measures having equivalent effect shall be prohibited between Member States.’ The concept of quantitative restrictions is quite limited, since it only covers total import bans and import quotas.35 In contrast, the concept of measures of equivalent effect embraces a very wide range of non-tariff barriers to imports. The precise scope of this concept has given rise to a vast body of case law as well as an extensive quantity of literature.36 Accordingly, we cannot explore that issue here. Suffice it to say that any act which discriminates against imports in favour of domestic products—other than a tax or charge or quantitative restriction pure and simple—constitutes a measure of equivalent effect (eg discriminatory restrictions on sale, advertising, and promotion, public supply contracts etc). 30 Case 302/86 Commission v Denmark (returnable bottles) [1988] ECR 4607. Today, the Court would no doubt refer in addition to Art 11 TFEU which is cited in section 11.5.4.2 below. 31 Opinion of AG Jacobs in case Outokumpu (n 27) para 54 ff. 32 Outokumpu (n 27) paras 38 and 39. Similarly, in Case C-221/06 Frohnleiten v Austria [2007] ECR I-9643, a discriminatory charge on imported waste was held to run counter to Art 110. 33 34 Essent Network Noord (n 10) para 57. Essent Network Noord (n 10) para 57. 35 Case 2/73 Geddo [1973] ECR 865. 36 Barnard and Peers (n 14) p 73 ff, Craig and de Búrca (n 14), Peter Oliver et al, Free Movement of Goods in the European Union, 5th edn (Hart Publishing, 2010) passim, and Oliver and Martínez (n 14) p 333 ff.

11.4 Free movement of goods: Restrictions

207

Equally, the concept of measures of equivalent effect also applies to many non-tariff barriers, even if they treat imports in the same way as domestic products. An obvious example of a measure of equivalent effect may be seen in Campus Oil.37 That case concerned a requirement imposed by Ireland that importers of petroleum products purchase a fixed proportion of their needs from Whitegate, the one remaining Irish refinery at the time. Not surprisingly, the Court had no difficulty in finding that this measure constituted a measure of equivalent effect within the meaning of what is now Article 34 TFEU.38 The far more important but controversial aspect of the judgment concerns justification, which will be considered below in section 11.5. This ruling was confirmed in PreussenElektra.39 By German law, each electricity supplier was obliged to purchase electricity from renewable sources within its area of supply in Germany. As in Campus Oil, it was never in doubt that this constituted a measure of equivalent effect, the real issue being the question of justification (see section 11.5 below). What is more, the Court of Justice has just decided two very similar cases which also concern Article 34 TFEU and measures taken to boost the use of renewable energy: Essent Belgium40 and Ålands Vindkraft.41 The Essent case reached the Court first and, after the hearing had taken place before a Chamber of five judges, AG Bot delivered his Opinion. Several months later, Ålands Vindkraft was heard before the Grand Chamber consisting of fifteen judges (clearly, the Court had realized the importance of these cases). AG Bot then delivered his Opinion in the latter case, which delved into the issues in far greater depth than that in Essent. However, for some unknown reason, the judgment in Ålands Vindkraft was delivered first. Nevertheless, it is appropriate to begin by considering Essent Belgium. Essent Belgium relates to measures taken by the Flemish Region pursuant to the 2001 RES Directive42 and the 2003 Electricity Directive.43 The Flemish Region decreed that, to meet the quota of green energy which each electricity producer was required to meet under this legislation, those producers could only rely on green certificates relating to green energy produced within Flanders. The Flemish authorities were authorized to waive this requirement in certain cases, but they did not do so with regard to the quantities of green electricity which had been produced in other Member States and in Norway and which had then been imported into Flanders by Essent Belgium. As regards the imports from other Member States, Essent claimed that this scheme was contrary to Article 34 TFEU. As the reader will be aware, Norway is not in the EU, but it is one of three countries which are parties to the Agreement on the European

37

Case 72/83 Campus Oil v Minister of Industry and Energy [1984] ECR 2727 (Campus Oil). Campus Oil (n 37) paras 12–20. 39 Case C-379/98 PreussenElektra AG v Schhleswag AG [2001] ECR I-2099. 40 41 Essent Belgium (n 4). Ålands Vindkraft (n 5). 42 Directive 2001/77/EC of the European Parliament and of the Council of 27 September 2001 on the promotion of electricity from renewable energy sources in the internal electricity market [2001] OJ L 283/33 (2001 RES Directive). 43 Directive 2003/54/EC of the European Parliament and of the Council of 26 June 2003 concerning common rules for the internal market in electricity and repealing Directive 96/92/EC [2003] OJ L 176/37 (2003 Electricity Directive). 38

208

Free Movement of Goods

Economic Area (EEA) with the EU.44 Article 11 EEA in effect replicates Article 34 TFEU. With respect to the imports from Norway, it claimed that the scheme was in breach of Article 11 EEA. As the Court noted, the rule that green certificates were only granted for green energy produced in Flanders constituted a two-fold restriction on imports: most obviously, it prevented electricity producers from meeting their quotas by importing renewable energy; but in addition it prevented those producers from trading in green certificates which they would otherwise have acquired by importing renewable electricity. Unsurprisingly, the Court therefore found that the restriction of the aid to renewable energy produced in Flanders constituted a measure of equivalent effect within the meaning of what is now Article 34.45 By the same token, the same was held to apply to Article 11 EEA as regards imports from Norway. Ålands Vindkraft concerned a Swedish restriction which was essentially the same as that in the Flemish case. However, a particularly poignant fact in the Swedish case is worth noting: the green energy concerned originated in the Åland Islands, a region of Finland which is linked by cable to Sweden but not to the Finnish mainland. More importantly, the Ålands case relates to the most recent EU legislation, namely the 2009 RES Directive.46 A bizarre feature of this Directive is that, although Articles 1, 3(1), and 5(1) refer to final mandatory national targets for ‘final consumption’, the term ‘consumption’ is in effect defined in Article 5(3) to mean production! In other words, each Member State is required to produce a certain quantity of green energy, not to consume it. This in turn affects the nature of the support schemes which Member States are authorized to establish pursuant to Article 3(3). ‘Support schemes’ are defined in Article 2(k) to mean: ‘Any instrument, scheme or mechanism applied by a Member State or a group of Member States, that promotes the use of energy from renewable sources by reducing the cost of that energy, increasing the price at which it can be sold, or increasing, by means of a renewable energy obligation or otherwise, the volume of such energy purchased.’ The third subparagraph of Article 3(3) provides: ‘Without prejudice to Articles 87 and 88 of the Treaty [now Articles 107 and 108 TFEU], Member States shall have the right to decide, in accordance with Articles 5 to 11 of this Directive, to which extent they support energy from renewable sources which is produced in a different Member State.’ This provision must be read with recital 25 in the preamble to the Directive which, in so far as is relevant, states: In order to ensure the effectiveness of both measures of target compliance, ie national support schemes and cooperation mechanisms, it is essential that Member States are able to determine if and to what extent their national support schemes apply to energy from renewable sources produced in other Member States and to agree on this by applying the cooperation mechanisms provided for in this Directive. 44

Agreement on the European Economic Area [1994] OJ L 1/3. Essent Belgium (n 4) paras 77 and 83–88. 46 Directive 2009/28 on the promotion of the use of energy from renewable sources and amending and subsequently repealing Directives 2001/77/EC and 2003/30/EC [2009] OJ L 140 (2009 RES Directive). 45

11.4 Free movement of goods: Restrictions

209

For good measure, it should be explained that cooperation mechanisms are arrangements between Member States and with third countries for achieving the national targets set by the Directive (Article 3(3)(b)). There was some dispute in Ålands Vindkraft as to whether the third subparagraph of Article 3(3) of the 2009 RES Directive purported to authorize the Member States to make the grant of support for renewable energy conditional on its production within their territory. The Grand Chamber found that it did do so.47 In addition, it confirmed that, in view of Article 5 of the 2009 RES Directive, the renewable energy quotas must be regarded as production, not consumption quotas.48 As to Article 34 TFEU, the German and Swedish Governments and the Swedish Energy Agency argued that the Court need not even consider that provision, because the 2009 RES Directive exhaustively harmonized the rules on national support schemes for green energy production. According to consistent case law, where a matter is exhaustively harmonized by EU legislation, the Court will examine the legality of national measures in the light of that legislation, not Article 34.49 The Court dismissed that contention on the basis that the Directive did not harmonize those schemes exhaustively; on the contrary, it gave the Member States considerable leeway in this regard, especially as it allowed them the right to decide to what extent to support green energy produced in other Member States.50 The Court also held that such a support scheme constituted a measure of equivalent effect under Article 34 TFEU, essentially on the same grounds as in Essent Belgium.51 Finally, price controls are another category of measure which may have particular relevance to energy policy. Even before Keck,52 the Court treated restrictions of this type as sui generis in that they were only held to fall under Article 34 TFEU if they placed imported goods at a disadvantage.53 This was confirmed by Keck, since price controls are a form of ‘selling arrangement’ and are thus subject to a discrimination test.54 Thus in the recent case of LIBRO which concerned resale price maintenance for books, the Court stated that the legislation placed imported books at a disadvantage and therefore fell under Article 34 TFEU.55

11.4.5 Article 35 TFEU Article 35 prohibits quantitative restrictions and measures of equivalent effect on exports between Member States. The term ‘quantitative restriction’ bears the same meaning mutatis mutandis as under Article 34.56 In contrast, according to longstanding case law, the concept of measures of equivalent effect to quantitative restrictions on exports under Article 35 TFE applies only to measures which discriminate 47

48 Ålands Vindkraft (n 5) paras 39–54. Ålands Vindkraft (n 5) para 47. Case C-309/02 Radlberger v Land Baden-Württemberg [2004] ECR I-11763, para 53 ff, and cases cited there. 50 51 Ålands Vindkraft (n 5) paras 56–63. Ålands Vindkraft (n 5) paras 65–75. 52 Case C-267/91 Keck [1993] ECR I-6097. 53 See eg Cases 65/75 Tasca [1976] ECR 291 (minimum prices) and 82/77 van Tiggele [1978] ECR 25 (maximum prices); and Oliver et al (n 36) para 7.87 ff. 54 55 See Oliver et al (n 36) para 6.60 ff. Case C-531/07 LIBRO [2009] ECR I-3717, para 21. 56 See n 35 and accompanying text. 49

210

Free Movement of Goods

against goods intended for export in favour of those destined for the domestic market.57 Thus this concept is markedly narrower in scope than that of import restrictions under Article 34 TFEU: as mentioned in section 11.4.4 above, the latter provision also covers a number of measures which apply to imports and domestic products in the same way. This case law has recently been confirmed by the Grand Chamber in Gysbrechts and Santurel, which rejected the suggestion by the Advocate General in that case that the test should in effect be the same mutatis mutandis as that applicable under Article 34 TFEU.58 Admittedly, the latter judgment did make it clear that de facto discrimination would suffice, but that much was already plain from the earlier case law.59 A detailed discussion of the merits of this case law would be out of place here. Suffice it to say that, while some widening of the scope of Article 35 TFEU would be welcome,60 transposing the Dassonville test61 and its progeny lock, stock, and barrel to Article 35 TFEU would surely be too radical.62

11.5 Free movement of goods: Justification 11.5.1 General As already mentioned, even if a measure is caught by Article 34 or Article 35, it will be lawful if it can be shown to be justified. To qualify for justification, a measure must fulfil two conditions: it must be intended to attain one or more of the grounds of justification which are spelt out in Article 36 TFEU or which are recognized by the Court; and it must comply with the general principles governing justification. In effect, the second condition means that the measure must be proportionate and must not be a ‘means of arbitrary discrimination’. These two issues will now be considered in reverse order.

57 Cases 15/79 Groenveld [1979] ECR 3409, 155/80 Oebel [1981] ECR 1993, C-388/95 Belgium v Spain (Rioja) [2000] ECR I-3123. 58 Case C-205/07 Gysbrechts and Santurel [2008] ECR I-9947. 59 Case C-350/97 Monsees [1999] ECR I-2921. Nevertheless, some have suggested that this judgment introduces a major change into the case law: Anthony Dawes, ‘A Freedom Reborn? The New Yet Unclear Scope of Article 29 EC’ (2009) ELRev 639; Anne Rigaux, case note, Europe: Actualité du droit communautaire 28, February 2009. For more measured views, see Alexandre Defossez, ‘L’article 35 TCE. Histoire d’une divergence et d’une possible reconciliation’ (2009) CDE 409; Stéphane Rodrigues, ‘Chronique de jurisprudence communautaire: Marché intérieur’ (2009) CDE 217, 230, and Wulf-Henning Roth, ‘Case C-205/07, Lodewijk Gysbrechts, Santurel Inter BVBA, Judgment of the Court of Justice (Grand Chamber) of 16 December 2008’ (2010) CMLRev 47 (2), 509–20 (case note on Gysbrechts). 60 Peter-Christian Müller-Graff in von der Groeben/Schwarze, Kommentar zum EU-/EG-Vertrag, 6th edn (Baden-Baden: Nomos, 2003), commentary on Article 29 EC, para 21 ff, Wulf-Henning Roth, ‘Wettbewerb der Mitgliedstaaten oder Wettbewerb der Hersteller?’ (1995) 159 ZHR 78 and case note on Gysbrechts (n 59). 61 Case 8/74 Procureur du Roi v Dassonville [1974] ECR 837. As is well known, the Court held there that all measures taken by Member States which are ‘capable of hindering, directly or indirectly, actually or potentially trade between Member States’ are measures of equivalent effect to quantitative restrictions on imports under what is now Article 34 TFEU (at p 852). 62 Oliver et al (n 36) para 6.106 ff.

11.5 Free movement of goods: Justification

211

11.5.2 Proportionality 11.5.2.1 The classic test The classic proportionality test is spelt out in De Peijper, where it was held that a measure can only be justified under Article 36 TFEU insofar as it is necessary to ensure the protection envisaged by the ground of justification relied on.63 This ruling has been confirmed on countless occasions.64 Frequently, the Court identifies an alternative measure which would achieve the desired aim without affecting interstate trade to the same degree.65 From this test it follows that, where EU legislation lays down exhaustively the guarantees required to meet the objective in question, recourse to Article 36 TFEU is no longer justified.66 If it were otherwise, there would of course be no purpose in the Union adopting harmonizing legislation in the first place. The latter rule was reiterated by the Court in Campus Oil.67 The Court therefore considered EU legislation requiring Member States to maintain minimum stocks of crude oil and petroleum products as well as related legislation, as it stood at the material time.68 The Court concluded that, although this legislation provided ‘certain guarantees’ that deliveries from other Member States would be maintained in the event of a serious shortfall in supplies, Member States still had no ‘unconditional assurance’ that supplies would be maintained at a sufficient level in any event.69 If the same issue were to arise again today, it is by no means clear that it would be decided the same way. Leaving aside the differences between EU legislation in issue in Campus Oil and that in force now,70 the finding in Campus Oil on this issue went hand

63

Case 104/75 De Peijper [1976] ECR 613, paras 16–17. Eg Cases 13/78 Eggers v Freie Hansestadt Bremen [1978] ECR 1935, para 30; Campus Oil (n 37) para 44; C-192/01 Commission v Denmark (vitamins) [2003] ECR I-9693, para 45; C-212/03 Commission v France (homeopathic medicines) [2005] ECR I-4213, para 43; and C-141/07 Commission v Germany (hospital pharmacies) [2008] ECR I-6935, para 50. In Case C-110/05 Commission v Italy (trailers) [2009] ECR I-519, the Court ruled the contested measure to be justified despite the absence of any cogent evidence from the defendant Member State. This controversial ruling can only be regarded as an isolated exception to the rule. Fortunately, nothing indicates that the Court intended to reverse the rule, especially as the subsequent case law has continued to apply it: eg Case C-333/08 Commission v France (processing aids) [2010] ECR I-757. 65 Thus a ban of the sale of goods has been held to be disproportionate where a labelling requirement would suffice: eg Cases 120/78 Rewe-Zentral (Cassis de Dijon) [1979] ECR 649, para 13, 788/79 Gilli and Andres [1980] ECR 2071, para 7, and C-319/05 Commission v Germany (garlic capsules) [2007] ECR I-9811, para 95. 66 Cases 35/76 Simmenthal v Italian Minister of Finance [1976] ECR 1871, 5/77 Tedeschi v Denkavit [1977] ECR 1555, and 148/78 Ratti [1979] ECR 1629. 67 Campus Oil (n 37) para 27. 68 Council Directive 68/414/EEC of 20 December 1968 imposing an obligation on Member States of the EEC to maintain minimum stocks of crude oil and/or petroleum products [1968] OJ L 308/14; Council Directive 73/238/EEC of 24 July 1973 on measures to mitigate the effects of difficulties in the supply of crude oil and petroleum products [1973] OJ L 228/1; Council Decision 77/706 setting a Community target for a reduction in consumption in the event of difficulties of supply [1977] OJ L 292/9 and Council Decision 77/186 establishing a system of export licences to allow the monitoring of trade [1977] OJ L 61/23. 69 Campus Oil (n 37) para 31. 70 Council Directive 2009/119 imposing an obligation on Member States to maintain stocks of crude oil and/or petroleum products [2009] OJ L 265/9; see Angus Johnston and Guy Block, EU Energy Law 64

212

Free Movement of Goods

in hand with its ruling in the same case on the public policy and public security exceptions in Article 36 TFEU; and that ruling was unduly indulgent towards the Member States. Unsurprisingly, the Court itself has retreated from it, as will be explained at section 11.5.4.1 below. Before considering the grounds of justification of particular relevance to energy policy, it is helpful to consider three forms which the protean principle of proportionality has assumed in recent years.71

11.5.2.2 Failure by the public authority to direct its mind . . . In Commission v Austria (Brenner I), the Court found the contested measure to be disproportionate because the province of Tyrol had failed to direct its mind to the possibility that other, less restrictive measures would have sufficed to achieve the desired result (in casu, the reduction of air pollution to the maximum level allowed by EU legislation on air quality).72 With this deft footwork, the Court avoided the need to decide whether the measure was in fact more restrictive than necessary to achieve the requisite reduction in air pollution: it sufficed that the Austrian authorities had failed to address this question.

11.5.2.3 Failure to allow an adequate transitional period In Radlberger v Land Baden-Württemberg, legislation was held to run counter to the principle of proportionality on the grounds that traders were given insufficient time to adapt to it.73 Member States may have to provide for a transitional period sufficiently long for producers and distributors to adapt to the requirements of any legislation imposing new burdens on them. Such a period may also be required to allow importers to sell stocks of products that no longer comply with the amended requirements.74 No doubt, it would be otherwise in the case of a genuine emergency.

11.5.2.4 Consistency and coherence In the 1980s, when the antiquated English law on Sunday trading was challenged as being contrary to what is now Article 34 TFEU,75 the English statute concerned was riddled with anomalies. For instance, it was prohibited to sell the Bible on a Sunday, but (Oxford: Oxford University Press, 2012), chapter 9 ‘Introduction to Energy Security and Security of Supply’, para 256 ff. 71 Takis Tridimas has described the principle of proportionality as ‘by its nature flexible and opentextured’: The General Principles of EU Law, 2nd edn (Oxford: Oxford University Press, 2006) p 173. 72 Case C-320/03 Commission v Austria [2005] ECR I-9871, para 87 (Brenner I). 73 Case C-309/02 Radlberger v Land Baden-Württemberg [2004] ECR I-11763, paras 80–81; see also Cases C-463/01 Commission v Germany (packaging) [2004] ECR I-11705, paras 79–80; Brenner I (n 72) para 90. 74 Case C-114/04 Commission v Germany (phytopharmaceuticals) judgment of 14 July 2005, nyr, paras 36–37. 75 See eg Case C-145/88 Torfaen v B&Q [1989] ECR 3831.

11.5 Free movement of goods: Justification

213

the sale of pornographic magazines was permitted.76 Assuming of course that it fell within Article 34 TFEU in the first place, nothing in the case law of the Court of Justice at that time suggested that these anomalies in any way prevented the measure from being justified. Since Gambelli that has changed: the Court ruled there that a restriction on establishment and services (in casu in the betting sector) could not be justified unless it was ‘suitable for achieving’ the desired objectives and that it must serve to attain those objectives in a ‘consistent and systematic manner’.77 Not only has this principle been repeatedly confirmed within the context of establishment and services,78 but it has also been applied to goods in Commission v Portugal (tinted film).79 The latter case provides a clear illustration of how this rule operates. It concerned a prohibition imposed by Portugal on affixing tinted film to car windows, thereby making it considerably more difficult to see into cars from the outside. Portugal maintained that this measure was justified for the purposes of crime prevention and road safety (to enable the police to check whether seatbelts were being worn). However, the Court found that this argument was undermined by the fact that Portuguese law permitted the sale of cars fitted with tinted windows from the outset.80 While this case law is to be applauded, there is a clear danger of it being taken too far. It would be absurd if a restriction which would otherwise be fully justified on (say) health grounds were to be found unlawful in its entirety merely because of some minor and irrelevant anomaly. That concern no doubt explains Barnard’s assertion that the Court has become excessively rigorous in its scrutiny of justification.81 Having said that, it should be remembered that the ‘consistent and coherent’ rule is itself a manifestation of the principle of proportionality; and it would be manifestly disproportionate to rule an entire scheme to be unlawful simply because of a minor and irrelevant anomaly. In short, the principle of proportionality is a rule of common sense and the Court can be expected to apply it in a common sense fashion.

11.5.3 Arbitrary discrimination The requirement that a measure must not be a ‘means of arbitrary discrimination’ does not call for a lengthy discussion here. Suffice it to recall that ‘arbitrary discrimination’

76 See Anthony Arnull, ‘What shall do on Sunday?’ (1991) ELRev 112; Peter Oliver, ‘Sunday trading and Article 30 of the Treaty of Rome’ (1991) Industrial Law Journal 298. 77 Case C-243/01 Gambelli and Others [2003] ECR I-13031, para 67. The thrust of Gjermund Mathisen’s article, ‘Consistency and coherence as conditions for justification of Member State measures restricting free movement’ (2010) CMLRev 1021, is that this ‘consistent and systematic’ test is not really new; but not everyone will be convinced. 78 Cases C-500/06 Corporacíón Dermoestética [2008] ECR I-5785, para 39, C-169/07 Hartlauer [2009] ECR I-1721, para 55, and C-316/07 Stoss [2010] ECR I-8069, para 24. 79 Case C-265/06 Commission v Portugal (tinted film) [2008] ECR I-2245. 80 Commission v Portugal (n 79) para 43. 81 Catherine Barnard, ‘Derogations, justifications and the four freedoms: is state interest really protected?’ in Catherine Barnard and Okeoghene Odudu (eds), The Outer Limits of European Union Law (Oxford: Hart Publishing, 2009) p 273 ff.

214

Free Movement of Goods

means disparate treatment which is unwarranted.82 For example, in a celebrated case, it was held that a ban on the importation into the UK of goods of a sexual nature could not be justified on public morality grounds (or indeed any other grounds), when the same goods could be freely manufactured and sold within that Member State.83 Exceptionally, it may happen that disparate treatment is objectively justified.84 In that event, no arbitrary discrimination will occur.

11.5.4 The grounds of justification Now we come to the crux of the matter, namely the grounds of justification. Article 36 TFEU sets out a list of various grounds of justification, including public policy and public security. Since its landmark judgment in Cassis de Dijon,85 the Court of Justice has supplemented this list with a series of judge-made grounds known as the ‘mandatory requirements’.86 These include environmental protection.87 Three grounds come into play in the present context: public policy, public security and environmental protection. For reasons which will become apparent, it is appropriate to consider the first two together, before turning to environmental protection.

11.5.4.1 Public policy and public security In Campus Oil,88 Ireland claimed that the requirement that petroleum suppliers purchase a fixed proportion of their needs from the Whitegate refinery was justified for the purpose of ensuring essential supplies, particularly in the event of a crisis, although the country produced no crude oil of its own. At the same time, this requirement plainly served to keep the refinery in operation and was thus of economic benefit to Ireland. The Court began by stating that the public policy exception was ‘not pertinent’.89 Although this assertion was unsupported by any reasoning, the Court probably took the view that public policy, being so general in nature, is an exception of last resort which is only engaged where no other, more specific grounds of justification are available. This appears to be part of a general trend in the Court’s case law to apply

82 See, in relation to imports, Cases 152/78 Commission v France (alcohol advertising) [1980] ECR 2299 and 121/85 Conegate v HM Customs and Excise Commissioners [1986] ECR 1007 and, in relation to exports, Case 53/76 Bouhelier [1977] ECR 197. 83 Conegate (n 82). 84 See Case 4/75 Rewe-Zentralfinanz v Landwirtschaftskammer [1975] ECR 843. 85 Cassis de Dijon (n 65). 86 This is not the place to debate the status of the ‘mandatory requirements’, ie whether they can only justify ‘indistinctly applicable’ measures (the traditional view which harks back to Case 113/80 Commission v Ireland (souvenirs) [1982] ECR 1625) or whether it can also justify ‘distinctly applicable’ measures (as in Brenner I (n 72) and Gysbrechts (n 58)). It is most heartening to find AG Bot robustly advocating the latter, progressive view in Essent Belgium (n 4) para 87 ff, and again in Ålands Vindkraft, para 79. That is also the view which is taken in Oliver et al (n 36) para 8.04 ff. 87 88 89 Commission v Denmark (n 30). Campus Oil (n 37). Campus Oil (n 37) para 33.

11.5 Free movement of goods: Justification

215

other grounds of justification wherever possible.90 In any case, the Court went on to declare that: [P]etroleum products, because of their exceptional importance as an energy source in the modern economy, are of fundamental importance for a country’s existence since not only its economy but above all its institutions, its essential public services and even the survival of its inhabitants depend upon them. An interruption of supplies of petroleum products, with the resultant dangers to the country’s existence, could therefore seriously affect the public security that Article 36 allows States to protect.91

It is settled law that Article 36 TFEU does not extend to purely economic objectives.92 After confirming this principle, the Court in Campus Oil nevertheless found that, in the light of the seriousness of the consequences that an interruption in supplies of petroleum products might have for a country’s existence, the aim of ensuring a minimum supply of petroleum products at all times transcended purely economic considerations. This aim was therefore capable of constituting an objective covered by the concept of public security.93 If the measures in issue were justified by ‘objective circumstances corresponding to the needs of public security’, it was of no consequence that those measures also made it possible to achieve ‘objectives of an economic nature’ (ie the maintenance of the Whitegate refinery).94 Following a consideration as to whether the measure in issue was proportionate, the Court concluded that: A Member State which is totally or almost totally dependent on imports for its supplies of petroleum products may rely on grounds of public security within the meaning of Article 36 of the Treaty for the purpose of requiring importers to cover a certain proportion of their needs by purchases from a refinery situated in its territory at prices fixed by the competent minister on the basis of the costs incurred in the operation of that refinery, if the production of the refinery cannot be freely disposed of at competitive prices on the market concerned. The quantities of petroleum products covered by such a system must not exceed the minimum supply without which the public security of the State concerned would be affected or the level of production necessary to keep the refinery’s production capacity available in the event of a crisis and to enable it to continue to refine at all times the crude oil for the supply of which the State concerned has entered into long-term contracts.95

Few lawyers would contest the proposition that measures taken to ensure the maintenance of essential supplies will fall under the public security exception in Article 36 90

See generally Müller-Graff in von der Groeben and Schwarze (n 60), commentary on Article 30 EC (now Article 36 TFEU) paras 49–54. 91 Campus Oil (n 37) para 34. 92 Cases 7/61 Commission v Italy (pork) [1961] ECR 317, 329; 95/81 Commission v Italy (import deposits) [1982] ECR 2187; 288/83 Commission v Ireland (Cyprus potatoes) [1985] ECR 1761; C-324/93 R v Secretary of State for the Home Department, Ex p Evans Medical [1995] ECR I-563 and C-254/98 Schutzverband gegen unlauteren Wettbewerb v TK-Heimdienst Sass [2000] ECR I-151, para 33. See also Wulf-Henning Roth, ‘Economic justifications and the internal market’ in Bulterman et al (eds), Views of European Law from the Mountain (Alphen aan den Rijn: Kluwer, 2009) pp 73–90. 93 94 Campus Oil (n 37) para 35. Campus Oil (n 37) para 36. 95 Campus Oil (n 37) para 51.

216

Free Movement of Goods

TFEU in appropriate cases. What is more, the Court stressed that such a measure was only justified in so far as it was strictly necessary to ensure the minimum required supply. Nevertheless, it was immediately clear that the Court had been unduly indulgent with Ireland in this case. AG Slynn had refrained from deciding whether the public security exception could ever justify a Member State with no crude oil of its own taking intrusive measures of this kind to keep a refinery in operation, given that the referring court in Ireland had yet to make a number of crucial findings of fact; but the Court did not regard this as any form of constraint. Moreover, it was, at the very least, arguable that a Member State with no significant crude oil supplies of its own would be equally well served by ensuring that its needs were met by having crude oil refined on the soil of a reliable ally. Accordingly, even at the time, a number of authors expressed surprise at the Court giving its blessing to such an apparently protectionist measure.96 More specifically, Marenco questioned whether the maintenance of a refinery in Ireland was necessary to refine oil covered by long-term contracts; in his view, Ireland could ensure supplies equally well by having such oil refined in refineries not situated on Irish soil.97 I wrote that the judgment in Campus Oil might be regarded as a serious blow to the prospects for creating a common energy policy.98 Perhaps that was hyperbole. In any case, the fact is that the Court has discreetly retreated from that ruling—a most welcome development. In the Greek Oil Supplies I case, the Court confirmed that ‘the aim of ensuring a minimum supply of petroleum products at all times is capable of constituting an objective covered by the concept of public security within the meaning of Article 36 of the EEC Treaty’.99 More specifically, in the Greek Oil Supplies II case the Court rightly confirmed that ‘the maintenance on national territory of a stock of petroleum products allowing continuity of supplies to be guaranteed constitutes a public security objective within the meaning of Article 36 of the Treaty’.100 No-one could sensibly quibble with those statements. Rather, it is the other findings in Campus Oil which are controversial and the Court has discreetly distanced itself from those, following some clear promptings from various Advocates-General. Arguably, this trend began with the Greek Oil Supplies I case. In the first place, the Court found that the State monopoly over imports of crude oil and petroleum products was not justified since Greece had failed to show that in the absence of this measure the public sector oil refineries would be unable to dispose of their products on the market at competitive prices and thereby ensure their continued operation.101 Second, it ruled to the same effect with regard to the obligation imposed on petroleum companies to

96 Laurence Gormley, Prohibiting Restrictions on Trade within the EEC (Amsterdam, North Holland, 1985) p 139; Giuiano Marenco, ‘La giurisprudenza communitaria sulle misure di effetto equivalente a una restrizionc quantitativa (1984–1986)’ (1988) Il Foro Italiano IV 166; and Peter Oliver, ‘A review of the case law of the Court of Justice on Articles 30 to 36 EEC in 1984’ CMLRev 22, 301, 307 ff. 97 98 Marenco (n 96). Oliver (n 96) p 312. 99 Case C-347/88 Commission v Greece (Greek Oil Supplies I) [1990] ECR I-4747, para 58. 100 Case C-398/98 Commission v Greece (Greek Oil Supplies II) [2001] ECR I-7915, para 29. 101 Greek Oil Supplies II (n 100) para 49.

11.5 Free movement of goods: Justification

217

notify their annual procurement programmes in advance.102 In particular, the Court found that it would suffice to require these companies to notify to the Greek authorities in due time their procurement programmes and any significant amendments made thereto in the course of their implementation.103 Having said that, the rulings on all these issues cannot necessarily be taken as a departure from Campus Oil, since that was a preliminary ruling and the referring Irish court was left to apply that ruling to the facts. The more plausible view is that the real turning point came with the Opinion of AG Cosmas in the Energy Import-Export Monopolies Cases.104 He began by defending the ruling in Campus Oil on the basis that the Court had struck the right balance: on the one hand, it had observed that an interruption of supplies of petroleum products, with the resultant dangers for the country’s existence, could seriously affect its public security; and on the other, the Court had insisted that the quantities of petroleum products covered by such a system must not in any circumstances exceed the minimum supply requirement without which the operation of the essential public services and the survival of the inhabitants of the State concerned would be affected.105 As explained earlier, this generous view of the judgment in Campus Oil is not universally shared. In any case, the Advocate General then added crucially: [ . . . ] If not only measures for ensuring the essential minimum, but also those which have the object of bringing about favourable conditions of supply from the viewpoint of cost, quality and selective management were regarded as measures designed to ensure the supply of a particular form of energy crucial to the functioning of the State and of the economy, there would be almost unlimited scope for potential exceptions to the principle of the free movement of goods. Any measure connected directly or indirectly with the conditions for production and marketing of the form of energy concerned and, finally, any measure covered by the relevant area of the energy policy of the Member State concerned could be considered to be related to grounds of ‘public security’. However, in practice that would be tantamount to setting apart energy policy as a special ground which could justify derogating from Articles [34 and 35] of the Treaty.106

Even though AG Cosmas defended Campus Oil, his Opinion can only be read as a warning of the consequences of applying the public security exception too broadly. The Court decided the cases on other grounds. Finally, the Greek Oil Supplies II case concerned Greek legislation which purported to implement EU Directives on the maintenance of minimum stocks of crude oil and/or petroleum products.107 The 102

Greek Oil Supplies II (n 100) para 58 ff. Greek Oil Supplies II (n 100) para 60. On the other points, the Court found for Greece. 104 Cases C-157/94 Commission v Netherlands [1997] ECR I-5699; Commission v Italy (n 10), C-159/94 Commission v France [1997] ECR I-5815 and C-160/94 Commission v Spain [1997] ECR I-5851 (Energy Import-Export Monopolies cases); see Johnston and Block (n 70) pp 243–4. 105 Joint Opinion of AG Cosmas in Energy Import-Export Monopolies cases (n 104) para 81. 106 Joint Opinion of AG Cosmas in Energy Import-Export Monopolies cases (n 104) para 82, emphasis in the original. 107 Council Directive 68/414 (n 68) as amended by Council Directive 72/425/EEC of 19 December 1972 amending the Council Directive of 20 December 1968 imposing an obligation on Member States of the EEC to maintain minimum stocks of crude oil and/or petroleum products [1972] OJ L 291/154. 103

218

Free Movement of Goods

Greek legislation required marketing companies either to keep such stocks themselves or to transfer that obligation to the refineries established in Greece by purchasing a significant part of their supplies from those refineries. Greece claimed that this scheme was justified under Article 36 TFEU so as to ensure the security of supply of petroleum products. The refineries’ fundamental right to economic freedom would be excessively restricted, it argued, if they were required to store the minimum stocks of petroleum products and thus assume an obligation of the marketing companies, unless the latter were required in return to purchase their supplies from those refineries. Taking its cue from the late AG Colomer,108 the Court dismissed this argument as being purely economic.109 In addition, the Advocate General found that less restrictive measures would have sufficed to protect public security.110 In particular, he rejected the contention advanced by Greece to the effect that, without the contested measure, it would be impossible to supply the armed forces with the special fuels which they used and which the marketing undertakings would be unable to sell to them: in his view, this measure was not essential since these fuels need not necessarily be produced or refined by national refineries.111 On these points, the Court also endorsed the Advocate General’s position, incorporating his reasoning by reference.112 Given the terms in which it was framed, it should have come as no surprise that the Court rejected Greece’s argument based on the refineries’ fundamental right to economic freedom as being purely economic. Nevertheless, it is striking that Greece’s argument was directly linked to its claim based on the need to ensure security of supply. Does this mean that the Court has reversed its finding in Campus Oil to the effect that the public security exception entitles a Member State with no significant crude oil supplies of its own to adopt some import restrictions—however limited—in the form of an obligation to purchase from the refinery or refineries situated on its territory so as to keep those refineries operating? In any case, in Essent Belgium113 AG Bot dismissed the Flemish Government’s claim that the measure contested there—which was very different from those in issue in the cases just mentioned—was justified to ensure energy supplies.114 As regards conventional sources of energy originating in third countries, there was manifestly no shortage of supply. What is more, there was no evidence of any risk to ‘internal’ supplies. The Court decided the case on other grounds.

11.5.4.2 Environmental protection As we noticed earlier, although environmental protection is not one of the grounds of justification mentioned in Article 36 TFEU, it was accepted as a ‘mandatory requirement’ capable of justifying restrictions on Article 34 and 35 TFEU in Commission v 108 109 110 111 112 114

Opinion of AG Colomer in Greek Oil Supplies II (n 100) para 43 ff. Opinion of AG Colomer in Greek Oil Supplies II (n 100) para 30. Opinion of AG Colomer in Greek Oil Supplies II (n 100) para 43 ff. Opinion of AG Colomer in Greek Oil Supplies II (n 100) para 46. 113 Greek Oil Supplies II (n 100) para 31. Essent Belgium (n 4). Essent Belgium (n 4) para 106.

11.5 Free movement of goods: Justification

219

Denmark (returnable bottles).115 What is more, Article 3(3) of TEU now states that attaining ‘a high level of protection and improvement of the quality of the environment’ is one of the objectives of the Union. Similarly, the ‘mainstreaming provision’ in Article 11 TFEU provides: ‘Environmental protection requirements must be integrated into the definition and implementation of the Community policies and activities, in particular with a view to promoting sustainable development.’116 As already mentioned, the measure in issue in PreussenElektra117 concerned a German law requiring each electricity supplier to purchase electricity from renewable sources within its area of supply in Germany. The Court noted that the growth in the use of renewable energy sources was among the environmental objectives of several international instruments and Community decisions. After that, the Court asserted that that policy was also designed to protect the health and life of humans, animals and plants; amongst the grounds of justification spelt out in Article 36 are ‘the protection of health and life of humans, animals or plants’. Next, it recalled the ‘mainstreaming provision’ in Article 11 TFEU (ex Article 6 the Treaty establishing the European Community) which has just been quoted. Furthermore, the Court noted that the recitals to the 1996 Electricity Directive118 referred to the need to give priority to the production of electricity from renewable sources.119 None of these general statements gives rise to any controversy. However, the Court then turned to specific considerations. ‘The nature of electricity’, it held, ‘is such that, once it has been allowed into the transmission or distribution system, it is difficult to determine its origin and in particular the source of energy from which it is produced’.120 Finally, the Court added its support to the view expressed by the Commission in its proposal for the 2001 RES Directive,121 that ‘the implementation in each Member State of a system of certificates of origin for electricity produced from renewable sources, capable of being the subject of mutual recognition, was essential in order to make trade in that type of electricity both reliable and possible in practice’.122 From these considerations the Court deduced that, ‘in the current state of Community law concerning the electricity market,’ legislation such as that in issue was ‘not incompatible’ with Article 34 of the Treaty.123 This reasoning was based on the observation that, once electricity has been allowed into the transmission or distribution system, its origin and the source of energy from which it is produced are difficult to determine. This was said to justify the discriminatory rule obliging electricity suppliers to purchase all electricity produced from

115

Commission v Denmark (returnable bottles) (n 30). For a detailed analysis of this provision, see Nicolas De Sadeleer, EU Environmental Law and the Internal Market (Oxford University Press, 2014), passim and Jan Jans and Hans Vedder, European Environmental Law, 3rd edn (Groningen: Europa Law Publishing, 2008) p 16 ff. 117 118 PreussenElektra (n 39). 2003 Electricity Directive (n 43). 119 PreussenElektra (n 39) paras 72, 75, 77–8. 120 PreussenElektra (n 39) para 79. 121 Proposal for a Directive of the European Parliament and of the Council on the promotion of electricity from renewable energy sources in the internal electricity market, COM/2000/0279 final [2000] OJ C 311E, pp 320–7 (Commission’s proposal for the 2001 RES Directive). 122 Commission’s proposal for the 2001 RES Directive (n 121) para 80. 123 PreussenElektra (n 39) paras 79–81. 116

220

Free Movement of Goods

renewable energy sources from producers of renewable energy within the respective supply area. However, that conclusion is difficult to reconcile with the principle of mutual recognition:124 the German provision in question did not even give importers of renewable energy the opportunity to provide a certificate from the Member State of production proving that the electricity was produced from renewable sources. The Court did not follow the approach which had been suggested by AG Jacobs in the same case. In his view, it was hard to see why ‘electricity from renewable resources produced in another Member State would not contribute to the reduction of gas emissions in Germany to the same extent as electricity from renewable resources produced in Germany’.125 The discrepancy between the Court’s position in PreussenElektra and the far more robust stance which it had taken a few years earlier in Outokumpu126 has not gone unnoticed.127 As already mentioned, in that case, the fact that Finnish legislation did ‘not even give the importer the opportunity of demonstrating that the electricity imported by him has been produced by a particular method’ was found to be fatal to the Finnish Government’s attempts to justify discriminatory internal taxation in that case.128 In any event, the exception relating to environmental protection has come to the fore again with regard to renewable energy in Essent Belgium129 and Ålands Vindkraft.130 The facts have already been set out. In the first of these cases, AG Bot took the view that PreussenElektra was no longer good law in view of the fact that new EU legislation had been adopted in the meantime.131 In particular, the very purpose of the 2001 RES Directive132 was in his view to guarantee that electricity produced from a renewable source in one Member State could be recognized as such in another, even though it was indistinguishable from energy produced from other sources once it was fed into the grid.133 He also endorsed the view expressed by AG Jacobs in the passage just cited from his Opinion in PreussenElektra.134 In addition, AG Bot dismissed the Commission’s analogy with the Walloon waste judgment where the Court had held that restrictions on imports of waste were justified by the principle enshrined in what is now Article 191(2) TFEU that environmental damage should where possible be remedied at source.135 In Ålands Vindkraft,136 after examining the issues at considerably greater length, AG Bot confirmed his position in Essent Belgium and rejected all the additional arguments

124 See eg Cassis de Dijon (n 65) and Case 272/80 Frans-Nederlandse Maatschappij voor Biologische Producten [1981] ECR 3277. See Markus Möstl, ‘Preconditions and limits of mutual recognition’ (2010) CMLRev 47, 405 and Oliver et al (n 36) paras 8.35 ff. 125 126 Opinion of AG Jacobs in PreussenElektra (n 39) para 236. Outokumpu (n 27). 127 Baquero Cruz and Castillo de la Torre, case note on PreussenElektra (2001) ELRev 26, 489. 128 129 Outokumpu (n 27) para 39. Essent Belgium (n 4). 130 131 Ålands Vindkraft (n 5). Essent Belgium (n 4) paras 102 and 103. 132 133 2001 RES Directive (n 42). Essent Belgium (n 4) para 103. 134 See the text accompanying n 125 above. 135 Case C-2/90 Commission v Belgium [1992] ECR I-4431, para 34. See Opinion of AG Bot in Essent Belgium (n 4) para 105. 136 Ålands Vindkraft (n 5).

11.5 Free movement of goods: Justification

221

advanced by the defendant Swedish energy agency and the intervening governments, purportedly under the general heading of environmental protection. First of all, it was argued that the support schemes would be undermined if Member States were precluded from confining their scope to renewable energy produced within their territory. Without such a condition, it was claimed, the Member States would be unable to control the effects and the costs of their support schemes. The Advocate General dismissed this contention, saying that mechanisms could be found to achieve this.137 Second, he rejected the claim that interstate trade in green energy required the prior conclusion of cooperation agreements between the Member States concerned. In his view, this argument was based on a misconception of the purpose of the cooperation agreements envisaged by the 2009 RES Directive, which was in fact to enable each Member State to reach its renewable energy target with the assistance of other Member States.138 Third, the Advocate General dismissed the argument that without the contested condition the Member States would lose their control over their choice between different energy sources and the general structure of their energy supply despite the second subparagraph of Article 194(2) TFEU. As he pointed out, that subparagraph only states that measures adopted under the common energy policy may not encroach on those national policies, but is expressed to be subject to Article 192(2)(c) TFEU— which makes specific provision for the adoption of legislation covering those very matters under the auspices of EU environmental policy. The 2009 RES Directive itself was just such a legislative enactment.139 Fourth, it was argued that there was a risk of ‘state aid shopping’. However, AG Bot found that this objection could not stand in view of the mechanisms for cooperation between Member States laid down in the 2009 RES Directive.140 The fifth and final argument was to the effect that opening national support schemes to energy produced outside the Member State concerned would require that State to finance the production of green energy in other Member States, which would be inappropriate. AG Bot gave this suggestion particularly short shrift: in his view, such considerations ran counter to the very principle of environmental protection which justified the consumers of a Member State giving financial support to green energy imported from another Member State in preference to energy produced from national fossil fuels.141 In conclusion, he pointed out that among the objectives which EU environmental policy is required to pursue according to Article 191(1) TFEU is the ‘prudent and rational utilisation of national resources’. In the Advocate General’s view, the sharing of renewable energy between Member States contributed to this objective.142

137 138 139 140 141 142

Opinion of AG Bot in Ålands Vindkraft (n 5) paras 96–8. Opinion of AG Bot in Ålands Vindkraft (n 5) paras 99–102. Opinion of AG Bot in Ålands Vindkraft (n 5) paras 103–4. Opinion of AG Bot in Ålands Vindkraft (n 5) paras 105–6. Opinion of AG Bot in Ålands Vindkraft (n 5) para 107. Opinion of AG Bot in Ålands Vindkraft (n 5) para 109.

222

Free Movement of Goods

For all these reasons, he found that the contested Swedish measure was not justified and was therefore contrary to Article 34 TFEU, as was the third subparagraph of Article 3(3) of the 2009 RES Directive, which purported to authorize the Member States to adopt this measure. However, he proposed that the validity of the latter provision should be maintained for a period of two years from the date of the Court’s judgment so as to allow the EU legislator to amend the Directive while avoiding a disruption in the market for renewable energy.143 That might well have been an empty gesture: how could the temporary validity of Article 3(3) of the Directive have had any practical effect when the national implementing measures would have been contrary to Article 34 TFEU in any case? In the event, this question does not arise, because the Court followed a radically different approach to justification. While acknowledging that the relevant EU legislation had changed substantially since PreussenElektra, it parted company from the Advocate General as regards guarantees of origin: the Court took the view that such guarantees when delivered by another Member State were not sufficiently reliable.144 The Court read the provisions in the 2009 RES Directive on the guarantees of origin to mean that their sole purpose ‘is to reveal to customers the proportion of energy from renewable sources in an energy supplier’s energy mix’ and that these guarantees ‘must be distinguished from green certificates used in the context of national support schemes and . . . do not, of themselves, confer the right to participate in such schemes’.145 Is this reasoning fully convincing? Having regard to the fact that the principle of mutual recognition is applied to countless other economic sectors,146 it is not obvious why renewable energy should be different. Moreover, that principle in no way precludes a Member State from taking the appropriate action should there be any evidence of fraud. In any case, the Court went on to set out a number of other grounds showing, in its opinion, that the scheme was justified. In particular, it found that the EU legislator’s choice of production, rather than consumption quotas ‘can be explained, in particular, by the fact that the green nature of electricity relates only to its method of production and that, accordingly, it is primarily at the production stage that the environmental objectives in terms of the reduction of greenhouse gases can actually be pursued’.147 With respect, what the possible causal link could be between this fact and the choice of production quotas is not clear. For the same reason it is hard to see any force in another ground expressed by the Court in support of this choice, namely that ‘the starting points, the renewable energy potential and the energy mix of each Member State vary’.148 In short, in Ålands Vindkraft the Court sitting in Grand Chamber chose to abide by the controversial judgment in PreussenElektra rather than plot the more appropriate course persuasively suggested by AG Bot: a scheme involving consumption rather than production quotas would have been equally effective.

143 144 146 148

Opinion of AG Bot in Ålands Vindkraft (n 5) paras 112–121. 145 Ålands Vindkraft (n 5) para 87 ff. Ålands Vindkraft (n 5) paras 52 and 89. 147 See n 124 above. Ålands Vindkraft (n 5) para 95. Ålands Vindkraft (n 5) para 98.

11.6 Conclusion

223

Two months later, the judgment was delivered in Essent Belgium, which related to the earlier EU legislation, namely the 2001 RES Directive149 and the 2003 Electricity Directive.150 Unsurprisingly, it found once again that the contested measure was justified on environmental grounds.

11.6 Conclusion In its November 2013 Communication, the Commission poignantly remarked: ‘The level, timing and nature of public intervention and how to reconcile such intervention with the internal electricity market and the EU acquis are questions which are becoming urgent, particularly in view of the completion of the internal electricity market in 2014.’151 As is plain from ACER’s Report152 and chapter 2, the failure to complete the internal market in renewable energy has led to distortions of the market—which, as mentioned at the outset, is not in the least surprising. This appears to be hard to reconcile with the free movement of goods, one of the fundamental freedoms enshrined in the EU Treaties, and with the requirement in Article 194(1)(b) to ensure security of supplies in a spirit of solidarity, which was inserted into those Treaties only five years ago. In October 2014, the Commission released a Communication entitled ‘Progress towards Completing the Internal Energy Market’.153 The title of this document reveals its key message: progress has been achieved towards this key goal, but it has not yet been met. Nor, of course, is this possible in the short time remaining before the end of 2014. Section 2.1 of this document is entitled ‘An integrated market is a basis for the cost-efficient decarbonisation of our energy systems’; and, as one would expect, this section goes on to mention renewables and to speak of the need to remove cross-border obstacles to trade in electricity generally. However, following the controversial judgments in Ålands Vindkraft and Essent Belgium, is there any incentive left for the Member States to complete the market in renewables?

149

150 2001 RES Directive (n 42). 2003 Electricity Directive (n 43). Communication from the Commission, Delivering the internal electricity market and making the most of public intervention, C(2013) 7243 final (November 2013 Communication) p 18. The European Council had already stressed the need to complete the internal market in electricity (and gas) by 2014. 152 ACER’s Report (n 6). 153 Communication from the Commission, Progress towards completing the Internal Energy Market, COM(2014) 634 final, 13 October 2014 (October 2014 Communication). 151

PART IV CASE STUDIES

12 Austria Thomas Starlinger and Harald Kröpfl

12.1 Introduction This chapter starts by outlining the current situation of electricity production in Austria, thereby setting the scene for an assessment whether or not a capacity mechanism, which is currently not established in Austria, might be necessary. It further evaluates the latest initiatives by the government with regard to capacity mechanisms, including the details of a draft act which was made public on 9 May 2014. By proposing a new set of state subsidies, the draft act aims to promote the ongoing operation of highly efficient cogeneration plants which are already in operation in Austria. This proposed capacity mechanism is analysed inter alia with regard to the EU law, in particular the EU rules on state aid.

12.2 Setting the scene 12.2.1 Market characteristics In the second half of the twentieth century the main part of Austria’s so-called hydrothermal network was established. Due to the special geography, conditions for hydro-energy were particularly advantageous: Run-of-river plants were built along the rivers while pumped storage power plants were constructed in the Alps. In addition, thermal power plants were built in the vicinity of populated areas as highly flexible support for the electricity system. Most of the pumped storage power plants, whose main purpose is to cope with peak loads and to store unused electricity when power generation would be high due to the strong current of the rivers, were commissioned by Verbund AG, Tiroler Wasserkraft AG, Vorarlberger Illwerke AG and KELAG with the participation of other local electricity suppliers.1 These undertakings are still majority state-owned, since electricity suppliers have to remain in public ownership by law,2 with at least 51 per cent of the share capital to be controlled either by the federal state or one of Austria’s provinces. The gross generation mix in 2012 showed that 65.7 per cent of the total electricity production in Austria was generated by hydro power; followed by thermal production, mainly gas and hard coal, with 30.5 per cent. ‘New renewables’ such as wind and photovoltaic only accounted for 3.6 per cent of the total power production, whereby 1

See APG, Masterplan 2030, p 12 ff. Federal Constitutional Law on the ownership of corporations of the Austrian electricity industry (Bundesverfassungsgesetz, mit dem die Eigentumsverhältnisse an den Unternehmen der österreichischen Elektrizitätswirtschaft geregelt werden), BGBl. I Nr. 143/1998. 2

228

Austria

wind power plants made up for 3.4 per cent. Thus, wind power provided the major part of the new renewable energy production in Austria whereas photovoltaic contributed only 0.2 per cent to the total power production.3 With two-thirds of the Austrian electricity production coming from green energy such as hydro and new renewables, Austria has a leading position in Europe with regard to ‘green power’. Compared to Germany, the new renewables however play only a minor part in the Austrian energy generation mix. This is due to the different approach Austria took with regard to the promotion of renewable energies. From the beginning of the promotion of the renewables with the Ökostromgesetz in 2002, the amount of subsidies aimed at the promotion of renewables was capped in Austria.4 Preferential feed-in tariffs for new renewables were fixed only for thirteen years rather than twenty years as in Germany. The difference between the wholesale price and the feed-in tariff is paid in the form of an eco-electricity subsidy. This approach avoided the excessive proliferation of wind and photovoltaic power plants that has taken place in Germany. The installed production capacity at the end of the year 2013 is proof of this: Hydropower plants amount to 13.4 GW and thermal power plants to 8.2 GW; whereas the total installed capacity for wind, photovoltaic, and geothermal only amounts to 2.1 GW. Since hydro power plants have a very stable power production, and thermal power plants and pump storage power plants are highly flexible, the Austrian power grid is not facing severe problems stemming from high fluctuations in power production from its own domestic market.5 Since the beginning of the liberalization of the electricity market in 2001 customers have the possibility to choose their electricity supplier. Hence, suppliers are in direct competition with each other. At the wholesale level of the market, the Energy Exchange Austria (EXAA) was established as an electricity exchange with access to the Austrian control area and all four German control areas. Since there are no congestions between the German and the Austrian control areas, German traders can also trade at the EXAA and there is one single clearing price for all control areas. Furthermore Austrian market participants have access to the EEX and EPEX Spot. The main electricity producers in Austria are Energie AG, EVN AG, Energie Steiermark AG, TIWAG-Tiroler Wasserkraft AG, Verbund AG, Wien Energie GmbH, Linz AG, KELAG, Illwerke AG, and Salzburg AG. Most of the electricity producers are incorporated as stock companies partly held by the state and by private companies like banks or other energy producers. Others, like for example Verbund AG, are listed on the Austrian stock exchange and have a significant part of their shares in free float. The main producers are organized under private corporate law. However, through direct or indirect participations, they are controlled by or at least under the influence of the state. Due to the aforementioned Federal Constitutional Law on the ownership of Austrian electricity companies, at least 51 per cent of

3

See E-Control, Key statistics 2014, report available at http://www.e-control.at/en/publications/ key-statistics, accessed 1 February 2015. 4 5 BGBl. I Nr. 149/2002. Jochen Homann, Energie Dialog (E-Control, 7 May 2013).

12.2 Setting the scene

229

the shares of the main electricity producers, as listed in the law, have to be controlled by the state.6 The majority part of the supra-regional high and extra-high voltage transmission system in Austria is operated by the Austrian TSO, Austrian Power Grid AG (APG). Besides that, since 2012, APG is the control area manager (Regelzonenführer) for the entire Austrian high voltage electricity grid. The entire transmission system is shown in Figure 12.1 below. The so-called 380-kV Ring is a project of APG to develop a closed 380-kV transmission system in Austria. For this project it is necessary to complete the 380-kV-Salzburg line (380-kV-Salzburgleitung), which is shown as a dotted line across the province of Salzburg. This new power line will close one of the last gaps in the 380-kV grid and is one of the most important infrastructure projects in Austria, replacing the 220-kV-line from St. Peter am Hart (Upper Austria) to the substation Tauern near Kaprun (Salzburg).7 Within the framework of the directives for European transmission grids, the 380-kV-Salzburg line was evaluated as being ‘of Community interest’.8 All cross-border interconnection points between Austria and its neighbouring countries, except for Germany, suffer from congestions, since these connections were not designed for the transportation of large quantities of electricity. Cross-border capacities, except those between Austria and Germany, are hence allocated by auctions.9 Compared with its total electricity consumption, Austria has a high exchange volume with neighbouring countries. In 2013, the domestic electricity consumption amounted to 69,912 GWh, while the exchange volume amounted to 42,649 GWh, with a decrease of the exports of 2,938 compared to 2011 (20,627 GWh). The amount of imports and exports to the neighbouring countries depends highly on the electricity prices in the respective country, Germany being the primary exchange partner. The imports from the Czech Republic far exceed the exports, while the opposite applies to Switzerland.10 However, this interconnection with the single continental European electricity grid is also causing major problems for the Austrian power grid and for the Austrian electricity market. This may be shown by the following two examples. First, on 30 November 2012 the APG-grid system had to deal with high hydropower production and therefore exports from Slovenia, Croatia, Bosnia, Serbia, Bulgaria, and Rumania as well as high imports to Italy and France which made massive interventions by APG necessary. Second, on the other side the APG-transmission system also had to handle critical north-south situations, as for example on 29 January 2013 and again on

6

Federal Constitutional Law (n 2). See http://www.apg.at/en/projects/380-kV-salzburg-line, accessed 1 February 2015. 8 Heinz Stigler et al, ‘Public interest regarding the establishment of the 380-kV-Salzburgleitung’ (Study for APG, September 2012), p 24 ff; Decision 1364/2006/EC of the European Parliament and of the Council of 6 September 2006 laying down guidelines for trans-European energy networks and repealing Decision 96/391/EC and Decision 1229/2003/EC [2006] OJ L 262/1; ENTSO-E, Ten-Year Network Development Plan (TYNDP) 2014, p 220, available at https://www.entsoe.eu/major-projects/ten-year-network-developmentplan/tyndp-2014/Pages/default.aspx, accessed 1 February 2015. 9 More information available at http://www.apg.at/en/market/cross-border-exchange/auctions, accessed 1 February 2015. 10 E-Control, Market Report 2014, p 33. 7

CZ Anlagen im Hoch- und Höchstspannungsnetz der Austrian Power Grid AG:

Slavetice

380-kV-Leitung

-Ri 380-kV ng

Projekt 380-kV-Salzburgleitung Pleinting Pirach

380-kV-Projekt Richtung Deutschland 220-kV-Leitung

BISAMBERG

Altheim Simbach

110-kV-Leitung

Ranshofen

geplantes Umspannwerk der APG APG Netzknoten

D

Memmingen Leupolz

Wagenham Hausruck

Ybbsfeld

ERNSTHOFEN

Salzach

Steweag-Steg

Ternitz

Weißenbach

H

Pongau

TAUERN

Hessenberg

WESTTIROL

Oststeiermark

Südburgenland

Steweag-Steg Steweag-Steg

Malta

CH Pradella

I

LIENZ

KAINACHTAL

OBERSIELACH

Maribor

Soverzene Podlog

SLO

Figure 12.1 Network of the Austrian Power Grid AG Source: APG11. 11

SARASDORF Neusiedl

Steweag-Steg

Bürs

WIEN SÜDOST

Salzburg

Energie AG

Zell am Ziller

SK

DÜRNROHR

ST. PETER

Umspannwerk der APG

Sokolnice

Use of figure was kindly granted by APG, illustration available at http://www.apg.at/de/netz/apg-netz, accessed 1 February 2015.

Györ Györ

12.2 Setting the scene

231

25 March 2013: On these days, due to a massive production of electricity at wind farms in northern Germany and lack of transmission capacity in Germany from north to south, the electricity flowed through the power lines of Poland to the Czech Republic and finally to Austria where the APG-grid had to cope with the surplus of electrical power.12 The electricity system and market in Austria is highly interlinked with the electricity system and market in Germany. As a consequence, the market prices in Austria are closely linked to market prices in Germany.13 The wholesale price has dropped from around €70 per MWh in 2007 to around €40 per MWh in 2013.14 Also the net export of electricity from Germany to Austria is growing very fast from around 6 TWh in 2010 to around 18 TWh in 2012.15

12.2.2 Regulatory framework Within the European regulatory framework the Austrian electricity business is subject to federal law, secondary legislation, and provincial law. At federal law level, the most important acts are the Electricity Act 2010 (the ElWOG 2010),16 the Act on the establishment of the regulatory authority for electricity and gas (the E-Controllaw),17 and the federal law regarding the coordination for ensuring security of supply (the Energy Intervention Powers Act 2012, Energielenkungsgesetz 2012).18 In accordance with the Austrian constitution each federal state has, in addition to the federal law and secondary legislation on federal level, its own electricity law which is only applicable in the respective federal state, but which has to conform with the framework given by federal provisions and federal secondary legislation.

12.2.2.1 Green Energy Act 2012 The promotion of electricity generation using renewable energies is regulated by the Green Energy Act 201219 which came into effect on 3 March 2011. From the beginning, the promotion of renewable prime energy sources has always been limited and certain caps were set. Hence, no major problems in granting network access to green energy currently exist in Austria. Green energy plants are promoted by way of purchasing the generated green electricity with special feed-in tariffs. These tariffs are guaranteed for

12

13 APG Masterplan 2030, p 48 ff. Österreichs Energie, February 2013, p 8. See Rainer Himmelfreundpointner, ‘Das große Strommarkt-Chaos’, online article of 7 May 2013, available at FORMAT.AT http://www.format.at/articles/1319/931/357943/das-strommarkt-chaos, accessed 1 February 2015. 15 Homann (n 5). 16 Federal Law BGBl. I Nr. 110/2010 (Elektrizitätswirtschafts- und –organisationsgesetz 2010, ElWOG 2010). 17 Federal Law BGBl. I Nr. 110/2010, BGBl. I Nr. 174/2013 (Bundesgesetz über die Regulierungsbehörde in der Elektrizitäts- und Erdgaswirtschaft, Energie-Control-Gesetz, E-ControlG). 18 Federal Law BGBl. I Nr. 41/2013 (Energielenkungsgesetz 2012, EnLG 2012). 19 Federal Law on the promotion of renewable energy BGBl. I Nr. 75/2011, as amended BGBl. I Nr. 11/2012 (Bundesgesetz über die Förderung der Elektrizitätserzeugung aus erneuerbaren Energieträgern, Ökostromgesetz 2012—ÖSG 2012). 14

232

Austria

thirteen years and are granted once a year on a first-come, first-served basis up to a certain ceiling. These purchases are executed by the Green Power Settlement Agency (the ‘Ökostromabwicklungsstelle’) which is obliged to buy electricity produced by renewable power plants up to the maximum amount of the total available aid. The consumer makes his contribution by paying a fee (the ‘Ökostromförderbeitrag’) in addition to the system charge. Furthermore, a lump sum (the ‘Ökostrompauschale’) has to be paid by consumers. On 18 September 2012, the latest Green Energy Feed-in Tariff Ordinance 2012 was published in the Federal Law Gazette, setting the current feed-in tariffs for green energy plants. A green energy plant is required to certify the produced electricity in order to have access to the benefits from the Green Energy Act. If the renewable power plant is located outside Austria, the power imported to Austria needs to be certified in accordance with the requirements of Regulation 2009/28/EC in order to qualify for the subsidies provided by the Green Energy Act. Once the electricity is fed into the electricity grid, it makes no difference if it is consumed in the Austrian market or exported to another country.

12.2.2.2 Energy Intervention Powers Act 2012 The Energy Intervention Powers Act 201220 gives the authorities the right to take intervention measures in case of any disruptions of Austrian energy supplies. Those powers may be invoked only if the failure of the power supply does not merely represent seasonal shortages and if it cannot be overcome by means of market-based measures. Intervention measures may also be taken if they are necessary to fulfil obligations under international law. Intervention measures shall be taken by order of the Federal Minister of Economy, Family and Youth with the approval of the Main Committee of the National Council. The Federal Minister has the power to enact far-reaching intervention measures, including issuing orders to generators, system operators, balance group coordinators and electricity wholesalers and retailers regarding the generation, transmission, distribution, and retailing of electricity as well as demand side interventions. Furthermore, the supply of electrical energy to other Member States could be regulated. The Federal Minister could also determine consumption quotas for the provinces. Intervention measures in relation to these consumption quotas however, have to be taken by the provincial governors. E-Control and the APG as the control area manager of the Austrian high-voltage grid, together with the distribution system operators (DSOs), which are control area managers of their respective distribution grids, and provincial authorities play a main role in preparing in advance a system which allows measures to be taken promptly when needed, in an efficient manner, whereby the control area managers are responsible for the operational implementation of the measures enacted by order of the Federal Minister.

20

EnLG 2012 (n 18).

12.2 Setting the scene

233

12.2.2.3 E-Control For carrying out the regulatory tasks in the area of electricity and gas E-Control, a public agency with own legal personality, was established by the law on the regulatory authority in the field of electricity and gas.21 E-Control has the task of strengthening competition, supervision and control of Third Party Access and network undertakings, while ensuring that this does not compromise security of supply and sustainability. To this effect, it is entrusted with the task of establishing market rules to enhance competition, including the power to determine network tariffs, regulate network tariffs, identify and remedy competition violations, track and analyse market developments, and the development of the market rules in consultation with market participants. The members of the decision making bodies of E-Control are independent and are not allowed to exercise any other function that might influence their independence. The Regulatory Commission of E-Control is competent to adopt orders determining the system charge, for decisions in connection with disputes caused by refusal of access to the electrical power grid and the settlement of other disputes between participants of the electrical power grid. E-Control is also in charge of the development of market rules and the assessment of technical and organizational rules for the operators and users of the electrical power system. Furthermore, E-Control prepares and publishes price comparison charts for end users and acts as their central information point. E-Control has to take precautions to ensure compliance with directly applicable EU-law and cooperates with the regulatory authorities in the other Member States to promote the development of a European energy wholesale market. The responsibility for the preparation and coordination of the intervention measures as provided by the Energy Intervention Powers Act 2012 is transferred to E-Control. The implementation of these measures is executed by APG as control area manager. In the case that a consumption quota has been introduced for the federal states, E-Control is entitled to ensure compliance with this quota with binding resolutions.

12.2.3 Generation adequacy At present, as Figure 12.2 illustrates, Austria has a high generation adequacy. There is no nuclear generation in Austria, as the construction of nuclear units is explicitly prohibited by law. Consequently, there is no risk of capacity gap from the phasing out of nuclear energy, something that Germany has to deal with.22 However, operators of thermal power plants (oil-, coal-, and gas-fired) face problems. Even though the fleet of thermal power plants in Austria is quite modern, generators had to devaluate their assets and even mothball most of their thermal plants. They were forced to take these steps due to low prices for electricity at the wholesale level which are mainly caused by an oversupply of electricity due to the high production levels of renewable energy in Germany. For instance, Wien Energie AG depreciated its gas-fired power plants nearly to zero and Verbund AG decided to mothball five of its thermal power plants, three of 21

E-ControlG (n 17).

22

Österreichs Energie, 02/2013, 36 ff.

234

Austria Leistungsm aximum der verfugbaren Kraftwerke Lastspitze

18000 16000 14000 12000

Prognose Kraftwerksle istung Szenario 1 Prognose Lastspitze Prognose Kraftwerksle istung Szenario 2

10000 8000 6000 4000 2000 0 1990

1994

1998

2002

2006

2010

2014

2018

Figure 12.2 Coverage of power consumption in Austria by domestic generation Source: E-Control, Monitoring Report 2013, p 9.23

them being located in Austria.24 The coal-fired power plant in Mellach, where the biggest and newest gas-fired power plant of Austria was built and which now will be mothballed, is still operating since the power plant is used to deliver long-distance heating to the capital of the province of Styria, Graz. The most recent available forecast for the generation adequacy in Austria until 2020 is very positive with a surplus of about 10 GW at a required safety margin of only 1.8 GW.25 However, this forecast did not include the previously mentioned recent reduction of thermal power capacities. Similarly, ENTSO-E’s Scenario Outlook and Adequacy Forecast 2014–2030 attests to a high generation adequacy in every given scenario in Austria.26 Notwithstanding the previously mentioned forecasts, from our point of view it is questionable whether the recent changes in the electriticy market could not, nevertheless, lead to a security of supply problem. As described earlier most of the operators of thermal power plants decided to shut their plants down or mothball them. New projects will not be realized. Since the margin between base-load and peak-load electricity has dropped significantly, pump storage power plants are no longer profitable enough to justify new projects. The business model of these power plants was that low cost base-load electricity is used to pump the water upwards and to release it to generate electricity at peak-load. Hence, flexible thermal power plants which are

Use of figure was kindly granted by E-Control. ORF, ‘Verwerfungen auf Strommarkt’, online article of 27 May 2014, available at http://orf.at/stories/ 2229915/2229911/, accessed 1 February 2015; Die Presse, ‘Gaskraftwerke bringen Wien fast 400 Mio. Euro Verlust’, online article of 18 December 2013, available at http://diepresse.com/home/wirtschaft/economist/ 1509211/Gaskraftwerke-bringen-Wien-fast-400-Mio-Euro-Verlust, accessed 1 February 2015; Der Standard, ‘Verbund schließt fünf Kraftwerke’, online article of 14 May 2014, available at http://derstandard.at/ 1399507448029/Verbund-schliesst-fuenf-Gaskraftwerke, accessed 1 February 2015. 25 E-Control, Monitoring Report 2013, p 8. 26 ENTSO-E, Scenario Outlook and Adequacy Forecast 2014–2030, June 2014 (ENTSO-E’s Report). 23 24

12.3 Energy-only market

235

currently barely needed are shut down while projects for new pump storage power plants are on hold. But without those sources of flexibility, it might be difficult to stabilize the highly fluctuating power generation from renewables in any possible situtation or scenario.

12.3 Energy-only market Austria has an energy-only market, that is, with no capacity mechanism. Yet, the existing system of an energy only market is causing major problems for most power plant operators. In our opinion, those problems are not appropriately addressed by the regulators. Given the current level of electricity prices on the central-western European wholesale market, most projects for establishing new gas-fired and pumped storage power plants in Austria are on hold.27 Several existing gas-fired power plants have been mothballed because they could not be operated economically as they did not even cover their variable costs anymore.28 Current market prices not only prevent investments in new power plants, in fact they are jeopardizing the profitability of existing generation units, resulting in decreasing operating hours of thermal power plants.29 Even the most modern and biggest thermal power plant in Austria, the combined gas cycle power plant Mellach, which had just been completed in 2012, is generating losses.30 Since there are no congestions between the Austrian and the German electricity grid, the EEX effectively is the price setter for the Austrian market. The EXAA plays only a minor role due to its marginal liquidity. In practice, the price development at the EXAA follows the prices at the EEX. As discussed earlier, Austria benefits from numerous pumped storage power plants which are able to store or release electricity power on demand. Because of the fluctuations in the production of energy from wind and photovoltaic, these power plants are severely needed to cope with the high volatility of the renewable energy production—which is subject to peak loads as well as standstills. But even the number of pumped storage power plants present in Austria will not be sufficient to stabilize the power grid if more gas-fired power plants are mothballed, and the rapid increase of renewables in Germany continues. Nonetheless, E-Control considers that Austria is sufficiently equipped with capacity reserves, that consequently, no capacity mechanism would be required, and that a

27 See the response of Österreichs E-Wirtschaft to the Commission’s 2012 consultation on generation adequacy (discussed in section 1.4.5) downloadable from the Commission’s website at http://ec.europa.eu/ energy/en/consultations/consultation-generation-adequacy-capacity-mechanisms-and-internal-marketelectricity, accessed 1 February 2015. 28 See comments of Leo Windtner, CEO of Energie AG, interviewed by Irmgard Kischko, ‘Desaster für die Energiepolitik. Interview: Der Chef der Energie AG fordert ein radikales Umdenken und einen massiven Netzausbau’ Kurier, 11 July 2013, available at http://kurier.at/wirtschaft/unternehmen/leo-windtnerdesaster-fuer-die-energiepolitik/18.671.593, accessed 1 February 2015. 29 Günther Strobl, ‘Verbund will Gaskraftwerk Mellach einmotten’, online article of 14 June 2013, available at http://derstandard.at/1371169523143/Verbund-Gaskraftwerke-am-Pruefstand, accessed 1 February 2015. See also the response of Wien Energie to the Commission’s 2012 consultation on generation adequacy (discussed in section 1.4.5) downloadable from the Commission’s website (n 27). 30 Strobl (n 29).

236

Austria

discussion on capacity mechanism would only make sense in a European context.31 According to E-Control, the main problem is the lack of European integration especially for the provision of sufficient balancing services. If some kind of capacity mechanism were to be introduced, E-Control pleads that this should occur on a European rather than on the national level like today, as this could lead to market distortions. Along the same lines as E-Control, WKO (Chamber of Commerce Austria) has argued32 that grid operators should ensure security of supply by using all possible sources of electricity generation. WKO rejects the notion of price regulations or price caps, but favours the idea that a capacity mechanism should be introduced by the Commission to ensure security of supply.

12.3.1 Recent efforts to establish an additional support scheme for electricity generation By ministerial proposal from the Federal Ministry of Science, Research and Economy, a draft for an Energy Efficiency Package of the federal state was submitted to the National Council on 9 May 2014.33 Part of this Energy Efficiency Package is a proposal for a federal law that aims to ensure the further operation of highly efficient cogeneration plants (KWK-Punkte Gesetz—the KPG). According to the draft, the benefits of the KPG would apply only to already existing cogeneration power plants. The KPG foresees the establishment of an association composed of the majority of the operators of cogeneration power plants, the so-called Branchenorganisation. The Branchenorganisation should decide on industry standards for a cogeneration-model (Branchenregeln). These Branchenregeln should determine the establishment of a transparency unit (Transparenzstelle). According to the KPG-draft, the Branchenregeln also have to foresee that consumers connected to the power grid should be obliged to buy so called KWK-Points (cogeneration points) independently from their consumption for each of their metering points. KWK-Points should function as a mere unit of measurement without having an intrinsic value. The amount of KWK-Points per metering point would depend on the network level34 the consumer is interconnected to. For the first period, a total of 71 million KWK-Points would be distributed among the cogeneration operators. For the following periods, the amount of KWK-Points would be adjusted according to the changes in the number of metering points. The allocation of the KWK-Points is planned as follows. The operators of the cogeneration power plants would forward data on their electricity production to the transparency unit and would then be entitled to an allocation of a corresponding 31 See E-Control at http://www.e-control.at/de/marktteilnehmer/news/themen-archiv/newsletter/ kapazitaetsmaerkte, accessed 1 February 2015. 32 Response of WKO to the Commission’s consultation on generation adequacy (discussed in section 1.4.5) downloadable from the Commission’s website (n 27). 33 Ministerial Proposal for an Energy Efficiency Package, 40/ME XXV. GP (Energieeffizienzpaket des Bundes), May 2014. 34 ElWOG 2010 (n 16) divides the electricity system into seven network levels, from the high voltage level one to the low voltage level seven.

12.4 European dimension

237

amount of KWK-Points. A number of KWK-Points to be ‘bought’ by the consumers would then be allocated to the operators. This number would correspond to the amount of electricity produced and fed into the power grid in relation to the total amount of electricity produced by all cogeneration plants in Austria. The Branchenregeln should determine the price for one KWK-Point. A price range from €0.50 to a maximum of €1 is set by the ministerial proposal. According to the ministerial proposal, the entry into force of the KPG would be subject to the approval of the Commission and have a fixed-term of four years. It remains to be seen whether or not this ministerial draft will pass the vote in the Austrian Parliament.

12.4 European dimension 12.4.1 Acknowledging the EU context In its Working Program 2018,35 the new Austrian federal government refers to European energy law and policy. The federal government aspires to influence European goals for energy policy by active involvement and intends to define the national goals in accordance with the framework set by European law. Moreover, the EU Energy Efficiency Directive36 is supposed to be implemented into national law within the new legislation period. Further major projects of the new government are the implementation of the Regulation on trans-European energy infrastructure37 and securing the construction of the 380-kV-Salzburgleitung. In its network development plan 2013, APG, the operator of the electricity transmission grid, also addresses developments on EUlevel regarding the intended energy mix, upcoming changes in legislation as well as future developments in neighbouring countries such as Germany and its nuclear phase-out, since these developments will have to be taken into account when taking decisions regarding the Austrian power grid. ‘Located in the geographical centre of Europe, Austria is massively confronted with these international developments. Congestion or misallocations in neighbouring transmission systems have an immediate impact on the Austrian system and will have to be dealt with directly’.38 Most of the references made in the political discussion refer to Germany and its excessive promotion of RES. As explained earlier, the rapid growth of renewables such as wind and solar power plants in Germany causes major problems for the Austrian power grid as well as for the operators of Austrian thermal and pump storage power plants as they generate losses or do not generate adequate returns on investment. 35 Work Programme of the Austrian Federal Government 2013–2018, Austria. A story of success (December 2013) p 32 ff. 36 Directive 2012/27/EU of the European Parliament and of the Council of 25 October 2012 on energy efficiency, amending Directives 2009/125/EC and 2010/30/EU and repealing Directives 2004/8/EC and 2006/32/E Commission Directive 2012/27/EU on energy efficiency [2012] OJ L 315/1. 37 Regulation 347/2013 on guidelines for trans-European energy infrastructure [2013] L 115/39 (Regulation on trans-European energy infrastructure). 38 AGP, Netzentwicklungsplan 2014.

238

Austria

12.4.2 Assessment of proposed capacity mechanism under EU law According to the Working Program 2018,39 a capacity mechanism to be introduced in Austria should take into account EU law, in particular the EU state aid rules. The capacity mechanism as proposed by the ministerial proposal of 9 May 2014, 40/ME XXV GP, was designed as a direct contribution from the users of the electricity grid to the owners of cogeneration plants. The proposal would force the end users to buy KWK-Points at a price determined by the Branchenregeln. The money raised would be directly transferred to the owners of the power plants. According to the explanation to the ministerial proposal 40/ME XXV GP, this approach was chosen in light of the practice of the Commission and the Courts40 that consider support measures not necessarily as public aid in the legal sense of Articles 107 and 108 TFEU. According to the reasoning of the ministerial proposal, the intended KWK-Points are not to be considered as intangible assets. This reasoning relies on past decisions of the Commission.41 However, the government’s interpretation of these decisions might not be correct. In Photovoltaics Romania, the Commission left it open whether or not a certificate of a certain amount of electricity being produced that is granted by the state is to be considered as state aid if the producer is allowed/obliged to sell this certificate. Rather, the Commission came to the conclusion that even if the measure had to be considered as state aid, it would be compatible with the internal market since the measure promoted renewable energy sources. This was seen as justification of the measure, if it would constitute state aid in the first place. In another decision42 regarding the NOx trading scheme between private companies set up by the Dutch state, this trading system was considered to be state aid since the NOx credits were provided for free and the state suffered foregone revenues. But again, the Commission concluded that such state aid is compatible with the common market since this scheme made a valuable contribution to the EU environmental policy. In yet another decision the Commission assessed the Combined Heat and Power (CHP) certificate system notified to the Commission by the Belgian government in 2004.43 These certificates serve as proof that a CHP installation achieved a saving of 1000 kWh in the respective year. Each supplier was obliged to show a certain amount of these certificates to the authorities each year and had to pay a fine if they did not reach the necessary amount. The producers of CHP could sell their certificates to other producers. The Commission concluded that there are two kinds of trading systems which have to be distinguished: (a) systems where the certificates have a market value and which the authorities could have sold, leading to foregone revenues and (b) systems where certificates are considered as official proof of a certain production that

39

Work Programme (n 35) p 32. See Cases N 608/2004 CHP certificates [2005] OJ C 240; SA.33134 Green certificates for promoting electricity from renewable sources (Photovoltaics Romania) [2011] OJ C 244; N 540/2000 French highways [2000] C 354; N 35/2003 The Netherlands (NOx Trading Scheme) [2003] OJ C 227/8; C-379/98 PreussenElektra AG v Schhleswag AG [2001] ECR I-2099; C-279/08 P Commission v Netherlands [2011] ECR I-07671. 41 In particular, see CHP certificates (n 40) and Photovoltaics Romania (n 40). 42 43 NOx Trading Scheme (n 40). CHP certificates (n 40). 40

12.4 European dimension

239

cannot be sold or auctioned to the recipient. The first kind of system is to be considered as state aid since the state suffers foregone revenues when offering the certificates for free; in the second kind of system the certificates have no value. The Commission considered the presence of aid only in the first kind of system. The ministerial proposal was carefully drafted against the background of the decision practice of the CJEU and the Commission. It clearly aims to ensure that the KPG falls outside of the scope of state aid rules. The KWK-Points shall be distributed by the Branchenorganisation, a private organization, consisting of the majority of the operators of cogeneration power plants. KWK-Points are just a certificate of proof to ensure the correct allocation of funds by the Branchenorganisation. Private households and entrepreneurs are to be obliged to buy these KWK-Points, independently from their power consumption; no government funds would be involved. The Transparenzstelle is supposed to implement the operation of the fund; no governmental organization would be involved. Hence, the chances are good that the KWK-Point system might not be considered as a form of state aid. But even if the KWK-Point system would be considered as state aid, it may be justified on the grounds that the promotion of the gasfired power plants is necessary to ensure the further expansion of the renewable energies, since KWK-Plants are necessary as stand-by units to balance the system. Another point put forward in the ministerial proposal is that the further operation of the cogeneration power plants is necessary to ensure that the heating systems of the major Austrian cities are not converted to use other primary energy sources which in turn might lead to an increase of air pollution in the cities. The Austrian co-generation power plants are very modern and fulfil very high emissions standards that might not be met by the average heating system of a household. According to section 3.4 of the EEAG 2014–2020,44 operational aid for high energy efficient cogeneration plants may be granted on the basis of the conditions applying to operational aid for electricity from renewable energy sources. In that regard, it is necessary that either the costs of producing electricity or heat exceed the market price or, in case of industrial use of the produced electricity and heat, that it can be shown that the production cost of one unit of energy using that technique exceeds the market price of one unit of conventional energy. In our opinion it is very likely that the ministerial proposal will be significantly amended since the Legal and Constitutional Service of the Federal Chancellery pointed out many concerns with regard to its compliance with the Austrian Constitution.45 With regard to the fundamental right to property it is seen as critical that a private organization, the Branchenorganisation, which is composed of the operator of the power plants who will benefit from the KWK-Point system, is entitled to determine the KWK-System. It is questionable if the delegation of tasks of the public authority to a private organization is in line with the constitutional law. Since no guidelines for the determination of the Branchenregeln are included in the KPG, the Branchenorganisation is totally free to set their own guidelines. Since the obligation of the customers to 44

Communication from the Commission, Guidelines on State aid for environmental protection and energy 2014–2020 [2014] OJ C 200/1–55 (EEAG 2014–2020). 45 Statement of 27 May 2014, 10/SN-40/ME XXV.GP.

240

Austria

buy KWK-Points interferes with the fundamental right to property, it has to be clarified why this measure is suitable, necessary, and proportional. In particular it is not clear why the obligation to purchase KWK-Points is linked to the metering point and not the consumption of the customer. Furthermore, the KPG sets no rules for the determination of the price for the KWK-Points, just a minimum and maximum. Hence, it is not possible to control the prices set by the Branchenregeln. Also the so called Transparenzstelle fulfills tasks of public authority but is established under private law. In light of the statement of the Legal and Constitutional Service of the Federal Chancellery, we expect that either the proposal of the KPG will be revised or, if the statement is ignored and the proposal comes into force, that it is very likely that it is going to be challenged at the Constitutional Court.

12.5 Conclusion At the moment, Austria has no capacity mechanism in place. The analysis of the current situation of electricity production capacities and the anticipated development of consumption in Austria by E-Control shows that the currently installed thermal generation capacities are not necessary from a security of supply perspective. However, it could be that even in the near future stand-by electricity production capacities will be necessary to cope with reduced production from renewables. Austria’s federal government put a proposal for a subsidization of highly efficient cogeneration plants forward. In our opinion this proposal is not to be considered as state aid but in any case would, in accordance with the 2014 environmental guidelines, fulfil the criteria for the treatment as compatible aid. However the proposal meets strong criticism, in particular with regard to its compliance with the Austrian constitution. It remains to be seen whether or not this proposal will pass the necessary vote in the Austrian parliament. Furthermore, it is very likely that the KPG will be challenged before the CJEU as it may be in breach of EU State aid rules.

13 Belgium Wim Vandenberghe and René Gonne

13.1 Introduction Concerns are frequently expressed by the Belgian government, the Regulatory Commission for Electricity and Gas (CREG, the energy regulator), the TSO (Elia), and other observers that there will be insufficient generating capacity to meet electricity demand in the coming years, especially in winter months. This risk to security of supply stems essentially from the nuclear phase-out, low investment appetite, retirement of existing plants, and intermittency of renewables. This motivates Belgium—like several other Member States—to develop capacity mechanisms. This chapter takes stock of the recently adopted legislation governing capacity mechanisms. Section 13.2 explains the present capacity situation in Belgium and the security of supply risks. Section 13.3 discusses the key elements of the newly introduced capacity mechanisms. Section 13.4 assesses whether the new reforms are compatible or not with EU law.

13.2 Setting the scene 13.2.1 Market characteristics Table 13.1 provides some key figures on electricity generation in Belgium which are relevant when assessing the need for capacity mechanisms.1 It shows that 57 per cent of electricity generation comes from nuclear power plants. The Nuclear Phase-Out Act2 not only contains a moratorium for the construction of new nuclear power plants, it also restricts the lifetime of all existing nuclear power plants to forty years.3 These lifetimes, however, can be extended by the government, notably to face security of supply concerns. This is precisely what the government has done with the Tihange 1 power plant (see section 13.2.3). 1

Regulatory Commission for Electricity and Gas (CREG), Annual Report 2013, pp 46–7. Other relevant facts and figures (including spark spread) can be found in CREG, Capacity remuneration mechanism, Study (F)121011-CDC-1182, 11 October 2012 (CREG’s study on capacity mechanisms) and CREG, Operation and price evolution on the Belgian wholesale electricity market—2013 Monitoring Report, Study (F)140430CDC-1319, 30 April 2014. See also Christophe Brognaux and Jonas Geerinck, Shaping a Vision for Belgium’s Power Landscape, The Boston Consulting Group, 24 June 2013. 2 Nuclear Phase-out Act of 31 January 2003, Belgian Official Gazette (Belgisch Staatsblad/Moniteur belge), 28 February 2003. 3 Nuclear power plants: Doel 1 (433 MW): 15 February 2015; Tihange 1 (962 MW): 1 October 2015; Doel 2 (433 MW): 1 December 2015; Doel 3 (1006 MW): 1 October 2022; Tihange 2 (1008 MW): 1 February 2023; Doel 4 (1038 MW): 1 July 2015; Tihange 3 (1046 MW): 1 September 2025.

242

Belgium Table 13.1 Electricity generation in Belgium, 2013 Primary energy

Generated power MWh

%

Nuclear Natural gas Coal Other self-generated power used locally Hydro (incl pump storage) Other

10.918 17.628 4.253 775 1.578 6.248

57.3 24.7 6.0 1.1 2.2 8.8

TOTAL

71.400

100

Source: Authors’ own table.

Since November 2010, Belgium has been part of the CWE region4 which enables the coupling of members’ markets through cross-border interconnections as well as linked power exchanges.5 The grid allows up to 3,500 MW of power to be imported into Belgium.6

13.2.2 Regulatory framework The Algemene Directie Energie/Administration de l’Energie (Federal Energy Administration), which is the ministerial energy department, is required to draft a forecast study every four years on anticipated electricity supply and demand and the expected security of supply (Forecast Study). The Forecast Study covers a time period of ten years and contains different working assumptions which vary in function of available capacity, retired plants, and new investments.7 The Forecast Study is drafted in cooperation with the Federaal Planbureau/Bureau fédéral du Plan (the country’s main government agency for economic, social and environmental planning) and in consultation with the energy regulator. A stakeholder consultation with Elia, industry and other stakeholders is held. The Forecast Study makes recommendations in order to safeguard security of supply, which the TSO must take into account when establishing its long-term grid development plan. One of the recommendations the Federal Energy Administration can make in its study is explicitly provided for in the Electricity Act, namely the use of state tendering for new capacities (see section 13.3.2.1 below).8 4

The Central and Western European (CWE) market comprises Germany, France, Austria, Belgium, the Netherlands, and Luxembourg. 5 There has been market coupling in place between Belgium, France, and the Netherlands since 2006, the so-called Trilateral Market Coupling. 6 This does not mean that this import capacity is available in its entirety at any one time as the generation capacity and power demand in neighbouring countries directly influence the import and export behaviour of these countries. 7 Article 3 of the Federal Act regarding the organization of the electricity market of 29 April 1999, Belgian Official Gazette, 11 May 1999 (Electricity Act). The most recent Forecast Study was published in 2009 (Forecast Study 2008–2017), available at http://economie.fgov.be/nl/modules/publications/general/ etude_perspectives_electricite_2008-2017.jsp, accessed 1 February 2015. A new Forecast Study is available in draft form at http://economie.fgov.be/nl/modules/publications/analyses_studies/etude_perspectives_ approvisionnement_electricite_horizon_2030_-_projet.jsp, accessed 1 February 2015. 8 Electricity Act (n 7) Art 3(2)(6).

13.2 Setting the scene

243

In terms of security of supply and guaranteeing grid stability, power stations are required to inform the Secretary of State for energy (energy minister) of planned closures and/or mothballing of stations, which have not been identified in the Forecast Study (see section 13.2.3 below).9 In addition, balancing generation and consumption is the primary responsibility of power producers and access responsible parties,10 which need to inform Elia of all the electricity exchanges it carries out (injections, offtakes, exchanges between access responsible parties, imports and exports). Elia is obliged to ensure real-time balance between generation and consumption in the Belgian control area, using energy reserves available under contract with generators and industrial customers in Belgium. Elia also needs to contribute to security of supply by ensuring adequate transmission capacity and network reliability.11 The CREG in turn is obliged to monitor investment in generation capacity especially in relation to security of supply.12

13.2.3 Generation adequacy Notwithstanding the satisfactory integration of Belgium into the CWE market (of which some countries actually have excess power capacities) and its interconnections, several studies carried out over the past few years indicate a risk of capacity shortage in the Belgian system mainly due to generation inadequacy.13 These generation inadequacy concerns have, in the past, centred on the effects of the planned nuclear phase-out. The phase-out would have meant that around 1800 MW of base-load capacity would no longer exist in 2015, which equates to 30 per cent of total nuclear capacity (see section 13.2.1 above). However, this nuclear security of supply concern has now been partially addressed by the government’s decision to extend the lifetime of one of the Tihange 1 nuclear power plants (962 MW) by an additional ten years.14 So Tihange 1’s lifetime has been extended from October 2015 to October 2025.15 Aside from this change, the phase-out schedule as set out earlier (section 13.2.1, at n 3) remains unaltered, meaning that by 2025 around 6 GW of nuclear generation will have been phased-out. The Nuclear Phase-out Act now also no longer contains the possibility for the government to extend the lifetime of some or all nuclear power plants on the basis of security of supply concerns. In other words, the phase-out schedule referred to in section 13.2.1 has become final.16 This addresses one of the 9

Electricity Act (n 7) Art 4 bis. Access responsible parties are transmission system users (eg generators, suppliers, traders) which ensure that supply and demand are in balance for the grid access point(s) concerned. 11 12 Electricity Act (n 7) Art 8. Electricity Act (n 7) Art 23. 13 CREG’s studies assessing Belgium’s security of supply since 2007 are available at http://www.creg.be/ nl/producte6.html, accessed 1 February 2015. See also a more recent study carried out by the Federal Energy Administration, Rapport sur les moyens de production d’électricité 2012–2017 (Final report, June 2012). 14 Law of 18 December 2013 amending the Nuclear Phase-out Act of 2003 and the Act of 11 April 2003 regarding the provisions for the dismantling of the nuclear power plants, Belgian Official Gazette, 24 December 2013. 15 The owners of Tihange 1 are obliged to pay an annual fee to the Belgian State in view of the lifetime extension of their plant. 16 Of course a nuclear power station might have to retire earlier, for example, due to technical problems which are considered a threat to the safety of the plan. See in that respect the reported problems with Doel 3 10

244

Belgium

issues voiced by investors in that the possibility provided for by the Nuclear Phase-out Act to extend the lifetime actually created legal uncertainty. Notwithstanding Tihange 1’s lifetime extension, two (other) nuclear power stations had to shut down from the summer of 2012 onwards due to alleged safety issues in the reactor vessels leading to a temporary reduction in generation capacity of around 2014 MW. This reduction was largely offset by a significant increase in net energy imports from both France and the Netherlands.17 The fact that Belgium was structurally dependent on imports during the entire winter period of 2012–2013 meant that it was unable to fulfill its usual role of being a transit country for electricity (to France). The CREG also takes the position that security of supply in Belgium cannot rely on net imports during extreme weather conditions.18 In addition, significant generation capacity from thermal production is being decommissioned and/or mothballed. This is largely due to the fact that more than half of the thermal production units are older than thirty and even forty years, and most no longer comply with stringent European19 and national emission limits and thus will need to be replaced.20 Aside from generation adequacy concerns and the increasing number of retirements of several conventional power plants, there is another type of security of supply risk. This is namely the lack of flexible generation capacity, especially in order to back up intermittent and ‘must run’ renewable resources from wind and solar.21 Concerns are expressed with regard to the ‘missing money’ problem22 and the lack of investment in new large-scale flexible gas-fired power generation. Support schemes indeed have generously promoted the integration of (intermittent) renewables in Belgium which in turn impacted the power market as the deployment of existing flexible technologies is no longer appropriately remunerated, leading to the ‘missing money’ problem. The lack of investment appetite also originates from the frequent and divergent changes to Belgian energy regulations (no stable legal framework) and the uncertainty for years about the precise calendar and terms of the nuclear phase-out.

and Tihange 2 in 2013 and 2014 which for now have led to a temporarily shut down of those nuclear units. See Federal Agency for Nuclear Control, Dossier Pressure Vessel Doel 3 & Tihange 2, available at http://www.fanc.fgov.be/nl/page/dossier-pressure-vessel-doel-3-tihange-2/1488.aspx?LG=2, accessed 1 February 2015. 17 Elia, System and Market Overview 2012, 1 August 2013, p 3. 18 CREG, Gas and electricity security of supply during lowest temperatures since market liberalisation— February 2012, Study (F)12 0801-CDC-1167, 1 August 2012 (Studie (F)12 0801-CDC-1167 van de bevoorrading szekerheid van aardgas en elektriciteit bij de laagste temperaturen sinds de vrijmaking van de markten—februari 2012). Also see the forecast analysis made by Elia in the context of the state reserve procedure launched in 2014 (Elia, Analyse van volume in het kader van strategische reserves—maart 2014, 20 March 2014). 19 Eg Directive 2001/80/EC of the European Parliament and of the Council of 23 October 2001 on the limitation of emissions of certain pollutants into the air from large combustion plants [2001] OJ L 309/1. 20 Forecast Study 2008–2017 (n 7). Also see, CREG, Installed generation capacity in Belgium in 2011 and its evolution, Study (F)111013-CDC-1113, 13 October 2011 (Studie (F) 111013-CDC-1113 over de geïnstalleerde capaciteit voor de productie van elektriciteit in België in 2010 en de evolutie ervan). 21 In total, 2185 MW of offshore wind capacity has been awarded. 22 For an explanation, go to sections 1.1 and 4.2.3.

13.3 Capacity mechanism

245

13.3 Capacity mechanism 13.3.1 Existing capacity mechanisms Broadly speaking, two types of capacity measures currently exist under Belgian law to address security of supply concerns in terms of generation adequacy and/or flexibility concerns. First, Article 5 of the Electricity Act foresees the possibility of tendering new power plants when security of supply is not guaranteed by power plants under construction, energy efficiency, and demand side measures.23 The energy minister needs to substantiate the use of the tendering procedure after having consulted Elia. Secondly, to address security of supply concerns, particularly in relation to the stability of the grid, the TSO can call upon reserve generation capacity.24 Reserve capacity is used to balance the electricity system and to assure system stability, amongst other functions. Reserve capacity is divided into three categories: primary, secondary, and tertiary reserve, which are activated depending on the type of imbalance. They differ in terms of their activation time and the duration of activation.25 The required volume of reserve capacity is assessed and calculated by Elia and approved by the CREG.26 The TSO contracts the reserve capacity in various ways: either from power plants which are technically sufficiently well equipped to provide this (production side measure); from industrial consumers which agree that a certain part of their offtake can be interrupted in specific cases of system vulnerability (demand side measure); or from other TSOs based abroad (eg RTE in France or TenneT in the Netherlands). This provision of reserve capacity leads to additional costs for Elia which are ultimately passed on to consumers through grid tariffs.27 Providers of tertiary reserve,28 for example, enter into capacity agreements with the TSO following successful participation in tenders organized by the latter. Providers receive remuneration for each MW capacity they make available per hour, even if the power plant is not producing (capacity payment). They receive further remuneration for each MW of produced power (energy payment). The CREG had suggested creating a fourth type of reserve (R4) which would give Elia special drawing rights during a limited period of time (when security of supply is at stake)

23 This is the transposition of the tendering option as stipulated in Article 8 of the Directive 2009/72/ EC of the European Parliament and of the Council of 13 July 2009 concerning common rules for the internal market in electricity and repealing Directive 2003/54/EC [2009] OJ L 211/55 (2009 Electricity Directive). 24 Electricity Act (n 7) Section 9. 25 Section 231 ff of the Technical Regulation for the management of and access to the transmission system dated 19 December 2002. Generally, the primary reserve can be activated very quickly (within zero to thirty seconds) and is used to maintain frequency; the secondary reserve can be activated within thirty seconds to fifteen minutes in response to typical imbalances; the tertiary reserve can be activated (fifteen minutes) in the event of major imbalances and substantial congestion. 26 Eg CREG, Decision (B)130626-CDC-1248 of 26 June 2013 regarding reserve capacity for 2014, available at http://www.creg.info/pdf/Beslissingen/B1248NL.pdf, accessed 1 February 2015. 27 Of course, a deviation from forecasted supply or demand by an energy market player will be charged to the latter through an imbalance tariff. 28 Elia (Grid support), Tertiary production reserve: a solution to major imbalances and congestion, available at http://www.elia.be/en/products-and-services/ancillary-services/purchase-of-ancillary-services, accessed 1 February 2015.

246

Belgium

from certain semi-base-load and peak-load power plants which have become unprofitable to operate.29 However, this measure was not pursued by the energy minister.

13.3.2 Proposed capacity mechanisms Considering this wider context of security of supply, the energy minister published his policy plan in June 2012 (Energy Policy Plan).30 This plan sets out high level actions needed to safeguard electricity security of supply in the short, medium and long term. The Belgian federal government approved the security of supply measures in the Energy Policy Plan on 5 July 2013.31 The Energy Policy Plan includes a series of measures to mitigate security of supply risks, including increasing cross-border interconnections32 and demand response.33 The Energy Policy Plan also includes new capacity mechanisms. The first one allows for state tendering for new capacities (section 13.4.1 below) and the second one creates a strategic reserve of dispatchable generating capacity (section 13.4.2 below). The discussion of these two capacity mechanisms is structured as follows. First, we provide a short description in the Energy Policy Plan. Then, we refer to the comments made by the CREG. Finally, we discuss the implementing legislation.

13.3.2.1 State tendering for new capacities Whether power stations are built or not depends on companies’ specific commercial and financial decisions. These decisions have been heavily influenced over the last few years by uncertainty in government policies (ranging from whether or not to phase-out nuclear power to subsidies for renewables). Additionally, the market outlook for flexible power plants and their profitability has been rather negative.34 In order to incentivize new-build flexible generation, the Energy Policy Plan proposes to tender 800 MW of capacity. The winning tender would obtain a guaranteed return on investment ‘for a certain period’. The exact yield percentage may differ in function of the ‘production level’ and in function of the ‘natural gas price and/or electricity price’. The tendering would be carried out on the basis of the existing Article 5 of the Belgian Electricity Act which is the transposition of the tendering option as stipulated 29 In other words, R4 would concern power plans which would be mothballed or closed. See CREG’s study on capacity mechanisms (n 1). 30 Le système électrique belge à la croisée des chemins: une nouvelle politique énergétique pour réussir la transition, 27 June 2015, also referred to as the ‘Plan Wathelet’, available at http://web.archive.org/web/ 20120916063556/http:/wathelet.belgium.be/wp-content/uploads/2012/07/Plan-Wathelet-pour-l%C3% A9lectricit%C3%A9.pdf, accessed 1 February 2015 (the Energy Policy Plan). 31 Council of Ministers, Press release regarding meeting of 5 July 2013, available at http://www. presscenter.org/nl/pressrelease/20130705/ministerraad-van-5-juli-2013, accessed 1 February 2015. 32 Belgium already has interconnections with France (available capacity between 2500 and 3200 MW) and the Netherlands (available capacity around 1400 MW). Currently proposed interconnections are with GB (project NEMO with a 1000 MW capacity) and with Germany (project Allegro with a 1000 MW capacity) by 2018. 33 Measures to reduce electricity consumption up to 1000 MW during peak demand (eg switch-off lights on highways) and additional controlled disconnection of large industrial users (70 MW). 34 Energy Policy Plan (n 30) p 27.

13.3 Capacity mechanism

247

in Article 8 of the 2009 Electricity Directive.35 The Energy Policy Plan does not provide any further details. The energy regulator has strongly criticized the state tendering option, particularly in terms of potentially discriminatory measures.36 First of all, according to the CREG, it is not demonstrated how the state tendering option can properly address security of supply risks which are identified for the period 2012–2017, as the construction and commissioning of power plants can take four years on average. Secondly, it is not demonstrated that all conditions are met to turn to state tendering (which should be a measure of last resort), notably as it is not shown that security of supply risks cannot be addressed through energy efficiency and/or demand side measures. Thirdly, no account is taken of the reduction in peak and total demand in the last five years. Fourthly, no consideration is given to the fact that, from 2018, there will be a substantial increase in interconnection capacities (projects NEMO and Allegro). Fifthly, the tendering is not technology-neutral and only applies to gas-fuelled power plants, and of these, it would seem that only CCGTs are eligible and not open cycle gas turbines (OCGTs).37 Sixthly, excluding OCGTs from this tender does not make technical and financial sense since OCGTs are more appropriate for peak demand (flexible), more profitable and less capital intensive than CCGTs. Finally, the tendering only applies to newly constructed installations and not to existing ones. The proposed measure of state tendering for new capacity is based on the existing Article 5 of the Electricity Act. For the first time, the government has now adopted decisions to implement this tendering option by way of executive orders (Ministerial Decision and Royal Decree).38 By doing so, and given the CREG’s critical response, the government has justified some of the design features but ultimately has not significantly altered its proposal as described in the Energy Policy Plan. The Ministerial Decision confirms essentially the need to launch a tendering process to respond to security of supply risks and summarizes some of the security of supply studies mentioned in section 13.2.3.39 Further, it states that there will be ‘a shortage in installed capacity in the Belgian market amounting to 2000 MW in 2017, and 4800 MW if imports are not taken into account’.40 The Decision itself (which consists of a mere two provisions) does not mention the scope of the tender in terms of required MW, though its preamble refers to the July 2013 decision of the Belgian government41 which stated that 800 MW is needed. All the Ministerial Decision does is determining the necessity to tender new capacity—this determination is the formal starting point of the tendering process which is in turn detailed in the Royal Decree. 35

2009 Electricity Directive (n 23). CREG, Opinion (F)130503-CDC-1243 regarding the conditions relating to the tendering process under Art 5 of the Electricity Act, 3 May 2013, available at http://www.creg.info/pdf/Adviezen/F1243NL.pdf, accessed 1 February 2015. In carrying out its analysis, CREG had privileged access to the draft tender specifications and the draft Royal Decree. 37 CREG, Opinion (n 36) pp 11–12. 38 Ministerial Decision dated 18 November 2013 concerning the use of the tendering procedure pursuant to Art 5(2) of the Electricity Act, Belgian Official Gazette, 2 December 2013 (Ministerial Decision); Royal Decree dated 8 December 2013 concerning the conditions of the tendering procedure under Art 5 of the Electricity Act, Belgian Official Gazette, 23 December 2013 (Royal Decree). 39 40 41 Listed at n 13. Ministerial Decision (n 38), Preamble. Council of Ministers (n 31). 36

248

Belgium

The Royal Decree is supposed to lay down the conditions of the tendering process for new capacity, which will be run by the Federal Energy Administration. However, the Royal Decree does not provide comprehensive information but rather general principles, which the Federal Energy Administration needs to take into account when drafting the tender specifications.42 Most of these are standard terms applicable in any public procurement process, such as the requirement that the tender contains selection and award criteria, steps and time limits for the different stages of the process, or confidentiality rules. There are however a couple of terms and conditions mentioned in the Royal Decree which are specific to the tendering for new generation capacity and which need to feature in the tender specifications to be drafted by the Federal Energy Administration.43 One of them is how much generation capacity (in MW) is requested through the tender process. Another is a description of the possible incentives.44 The tender should also mention whether it accepts a bid from only one power plant (which would provide the requested MW in its entirety) or rather several bids from multiple power plants. And finally, should more than one power plant be invited to bid, then separate rankings can be made in function of type in case they differ in technology (eg OCGT and CCGT). According to the Royal Decree, the Federal Energy Administration needs to make a substantiated review of the bid(s) and make a recommendation to the energy minister after having sought the opinion of the TSO regarding the bid(s).45 The energy minister then awards the contract to one or more bids by way of a ministerial decision.46 However, the energy minister may decide not to award any contract if this is justified by ‘important developments which have arisen after the call for tenders’.47 Following the adoption of the Ministerial Decision and the Royal Decree at the end of 2013, the Federal Energy Administration actually launched a tendering process on 15 January 2014.48 According to the tender specifications, the tender seeks a total capacity between 700 and 900 MW and is limited to power plants which are new (as opposed to existing capacity),49 located in Belgium or in a neighbouring country,50 and 42

43 Royal Decree (n 38) Art 3. Royal Decree (n 38) Art 3(4), 3(10), 5 and 15. It is thus left to the discretion of the Federal Energy Administration to decide whether or not to provide for incentives and if so, under which terms. 45 46 Royal Decree (n 38) Art 14 and 15. Royal Decree (n 38) Art 17. 47 Royal Decree (n 38) Art 16. 48 The tender specifications are available at http://economie.fgov.be/nl/ondernemingen/energie/ elektriciteit/Liberalisering_elektriciteitsmarkt/Elektriciteitsproductie/procedure_nieuwe_installaties/#. UwXvPoVnhrV, accessed 1 February 2015. The tender was also published (as required by EU law) in the European Official Journal at http://ted.europa.eu/udl?uri=TED:NOTICE:12340-2014:TEXT:EN:HTML, accessed on 1 February 2015. 49 Though see n 50 below with regard to (existing) power plants located abroad. 50 Reportedly in response to a question from Essent (RWE) in view of two of its power plants located on the Dutch-Belgian border (see Belgian newspaper, De Tijd, 21 February 2014), the Federal Energy Administration acknowledged that an existing power plant located abroad could participate in the tender under the following two conditions: (a) the capacity is entirely part of the Belgian operating reserve and (b) the capacity is only connected to the Belgian operating reserve. In that sense, an existing (foreign) power plant which would be connected to the Belgian transmission system after the launch of the tendering process, would qualify as a ‘new power plant’ in the meaning of the tender specifications. See Q&A related to the tender available at: http://economie.fgov.be/nl/ondernemingen/energie/elektriciteit/Liberalisering_ elektriciteitsmarkt/Elektriciteitsproductie/procedure_nieuwe_installaties/questions_reponses.jsp#. UwXvo4VnhrU, accessed 1 February 2015. 44

13.3 Capacity mechanism

249

exclusively gas-fuelled (OCGT or CCGT).51 Winning power plant(s) will be required to offer ancillary services in terms of balancing and frequency on the Belgian electricity market. Furthermore, the tender provides for a support mechanism under which each bidder can request the State for a subsidy payment per installed MW in order to guarantee the profitability of the plant for a certain duration. The bidder is free to determine the level of the subsidy payment requested though there is an absolute annual cap of 44,470.884 €/MW for an OCGT and 89,329.596 €/MW for a CCGT.52 The subsidy will be paid to the winning bidder(s) for a duration of six years from the start-up of the power plant. The subsidy is revised each year and can vary in function of three elements: (a) the effective annual generation of the power plant (MWh), (b) the positive clean spark spread for each hour that the power plant generated, and (c) the actual market circumstances compared to the reference models.53 Interested parties had to submit their bids by 22 July 2014.54 Following the closing of the submission process, the Federal Energy Administration has to seek the opinion of the TSO on the submitted bids and will then review and draw up a ranking of the bids before making a recommendation to the energy minister who is expected to take a decision in 2014 according to the timeline laid down in the Royal Decree.55

13.3.2.2 Strategic reserve Plant operators have announced in recent years the permanent closure or mothballing of a number of generation facilities. The Energy Policy Plan proposes the creation of a strategic reserve of dispatchable generating capacity from plants which are scheduled for closure or mothballing. It appears that this strategic reserve would not include existing (operational) or new generation capacity. In return for a remuneration, capacity providers of these out-of-market plants will be obliged to deliver energy at times of system stress and when called upon by Elia. Rather than a capacity auction, the strategic reserve and its remuneration level would be the result of negotiations between the TSO and the concerned plant operator 51

An OCGT has a minimum size of 40 MW, CCGT has a minimum size of 400 MW. Most likely in view of CREG’s criticisms, the choice for OCGT and CCGT only has been justified on p 2 of the tender specifications (n 48). 52 Installed MW. This maximum cap is based on reference models based on a range of assumptions— both for OCGT and for CCGT—in terms of: average CAPEX, annual fixed and variable costs, revenues and company taxation, amortization period, operational hours, clean spark spread etc. 53 See n 52. 54 According to press reports, three bids (by Delta/EDF; Essent/RWE; Dils Energie) were submitted of which two by power plants located in the Netherlands which would be connected to the Belgian transmission grid if the bid is successful. See: De Tijd, ‘Nederlandse centrales snellen België ter hulp’, 24 July 2014 http://www.tijd.be/nieuws/ondernemingen_energie/Nederlandse_centrales_snellen_Belgie_te_hulp.95264633088.art?ckc=1, accessed 1 February 2015. 55 The ranking of the bids (one for OCGT and one for CCGT in case bids with such differing technology are submitted) is done on the basis of the following award criteria which are specified further in the tender specifications of 2014 (n 48): (a) level of subsidy requested by the bidder (70% weighting), (b) date of commissioning of the power plant (15% weighting), (c) contribution to a competitive functioning of the generation market (10% weighting), (d) technical quality (5% weighting). Criterion (c) seems to indicate that companies which already have significant generation capacity in Belgium will score lower compared to new entrants or companies with fewer installed capacity.

250

Belgium

(subject to approval by the CREG). According to the Electricity Act, each plant operator is free to decide to close its plant or temporarily suspend its activities.56 The Energy Policy Plan considers different terms and conditions for each of the two scenarios. First, if a plant operator decides to close a plant, it must notify the Federal Energy Administration of its intention. Subsequently, an auction is organized to verify if third parties are interested in buying the plant. If yes, then the ownership of the plant is transferred and the plant remains in the energy-only market. If not, then the energy minister may take the decision to include the plant in the strategic reserve for a certain period of time and receives a compensation for this. This means that the plant operator should ensure the plant’s readiness and use, at least on the balancing market, and possibly also on the day-ahead and intraday markets. A contract is signed between Elia and the plant operator for an initial term of two years (which can be prolonged) which sets out the conditions to re-activate the plant and to ensure that the plant no longer participates as such in the energy-only market. Secondly, if a plant operator intends to mothball a plant, it needs to inform the Federal Energy Administration and the Energy Regulator and subsequently a contract is drawn up between the TSO and the operator to specify the conditions of the plant’s use in case of emergency situations during the period of mothballing (the contract term thus seems to be limited to this period). This measure is secondary to the use of the strategic reserve of closed power plants (ie measure of last resort). The CREG emphasized that in order for the strategic reserve to work properly, the following parameters should be defined: lifetime of the measure, maximum volume of the reserve, activation rules when procuring reserve capacity and penalties in case of non-compliance with the activation rules. These comments were made as part of the CREG’s study on capacity mechanisms,57 and not with regard to the Energy Policy Plan and/or implementing legislation. On 12 February 2014 the government proposed an amendment to the Electricity Act which would implement the strategic reserve measure (draft bill).58 In order to still have sufficient time to establish a strategic reserve for the period 2014–2015, the government requested the application of a special parliamentary approval procedure to hasten the draft bill’s progress. Subsequently, the draft bill became already law on 26 March 2014.59 In the explanatory memorandum attached to the draft bill, the government states that a strategic reserve is complementary to the state tendering for new capacity, as the former can be achieved in the short term whereas the latter takes several years.60 Elia is required annually to forecast the security of supply for the coming winter period.61 The forecast of the appropriate level of security of supply is made in reference to EU harmonized norms (possibly on a regional basis, eg CWE market). In case no

56

57 Electricity Act (n 7) Art 4a. CREG’s study on capacity mechanisms (n 1). Draft bill, Parl. St. 2013–2014, 53-3357/001. 59 Law dated 26 March 2014 amending the Electricity Act, Belgian Official Gazette, 1 April 2014 (Strategic Reserve Law). 60 61 Draft bill (n 58) pp 4–5. Electricity Act (n 7) Art 7bis. 58

13.3 Capacity mechanism

251

such norms are available, then the LOLE calculations are applied.62 Elements to take into account are generation and storage capacities, demand forecasts, and interconnection capacity. The Federal Energy Administration subsequently advises the energy minister about the necessity to create a strategic reserve, which can be established for a period of one year, two years, or three years depending on the expected level of security of supply.63 The Federal Energy Administration also makes a proposal of the required capacity volume (in MW). After receiving the advice, the energy minister can instruct the TSO to establish a strategic reserve.64 To procure the generating capacity as determined by the energy minister (in MW and total duration), Elia has to conduct a competitive tendering process. In doing so, Elia has to draw up and publish objective, transparent and non-discriminatory tender specifications after having first consulted the Federal Energy Administration, the Regulator, and network users.65 The TSO also draws up the operating rules which focus on the technical and administrative rules and procedures for how the strategic reserve will operate and aim at minimizing interference with the electricity market.66 These operating rules are subject to the CREG’s approval.67 Successful bidders will conclude an agreement with the TSO.68 Power plants eligible to participate in the tender are plants in the process of closing or mothballing, or which are effectively mothballed. Operators of these plants are actually under a duty to submit a bid, and are subject to penalties in case they do not do so.69 Power plants which are operational (ie not in the process of closing or mothballing) are not mentioned in the Strategic Reserve Law as being eligible. Aside from the power plants which are obliged to participate, the Strategic Reserve Law only refers to the possibility for transmission or distribution system users to propose demand capacity measures.70 Interconnected non-Belgian capacity and interconnectors seem to be implicitly excluded as bidders should ‘dispose of power within the Belgian electricity network’.71 Storage facilities are implicitly excluded as bidders need to ‘operate a generation facility’.72 Nuclear generation is explicitly excluded.73 Participants in the tender need to mention the level of payment they seek in their bid. If these proposed prices are ‘reasonable’ according to the energy regulator,74 the TSO may draw up agreements with the winning bidder (or bidders) after having received approval from the CREG. Should the energy regulator argue that the price in a bid (or bids) is ‘manifestly unreasonable’, then the energy minister has the power to set the price for the concerned bid.75 In addition, the energy minister may fix the volume of MW which can differ from the volume proposed in the bid. The expenditure related to the strategic reserve (paid by the System Operator to the successful bidders) will be covered by a public service obligation levy which will form part of the overall

62

63 See section 5.5.1 for an explanation of LOLE. Electricity Act (n 7) Art 7ter. 65 Electricity Act (n 7) Art 7quater. Electricity Act (n 7) Art 7quinquies. 66 67 Electricity Act (n 7) Art 7octies. Electricity Act (n 7) Art 7septies. 68 Electricity Act (n 7) Art 7sexies (3). 69 The mandatory placing of a plant in the strategic reserve as mentioned in the Energy Policy Plan (n 30) is not withheld in the Strategic Reserve Law (n 59). 70 71 Electricity Act (n 7) Art 7quinquies (2)(1). Electricity Act (n 7) Art 7quinquies (2). 72 73 Electricity Act (n 7) Art 7quinquies (2)(2)–(2)(4). Electricity Act (n 7) Art 7novies. 74 75 Electricity Act (n 7) Art 7sexies (2). Electricity Act (n 7) Art 7sexies (3). 64

252

Belgium

transmission tariffs.76 No further details are given in the Strategic Capacity Law with regard to the cost components nor by whom or when the exact amount of compensation will be decided (which in turn will need to correspond with the network tariff). The first tender to establish a strategic reserve was launched in 2014. Through a ministerial decision of 3 April 2014, Elia has been instructed to launch a tender to procure 800 MW annually over a period of three years starting 1 November 2014.77 An additional 400 MW for one year was requested by the energy minister following an update by the TSO of its forecast analysis of the security of supply.78 The key steps to implement the tender (as required by the Strategic Reserve Law and described earlier) have been taken. First, Elia published its security of supply forecast analysis.79 Next, the Federal Energy Administration gave its opinion.80 Elia subsequently established the procedure to establish the strategic reserve and the CREG gave its comments on it.81 Lastly, Elia developed the operating rules which were approved by the energy regulator.82 Aside from the establishment of a strategic reserve, the Strategic Reserve Law also contains an amendment of the rules on mothballing and closure of power plants during winter periods. Namely, the current Electricity Act contains in its Article 4bis an obligation on the part of the power plant operator to inform the energy minister, the CREG, and the TSO in case of a planned closure or mothballing. The proposed legislation modifies this duty to inform, in particular with regard to when information must be disclosed. This is done to ensure that decisions creating strategic reserves are based on the most recent available market information and developments. The proposed rules do, in effect, impact on the operator’s decision when to close or inactivate its power plant. In more detail, a plant operator is obliged to inform the energy minister, the CREG and the TSO about its decision to close or mothball a power plant by 31 July of the year prior to the date of mothballing/closure. Further, mothballing may only happen after 31 March and closure may only take place after 30 September following the year of 76

Electricity Act (n 7) Art 7octies. Ministerial Decision of 3 April 2014, available at http://economie.fgov.be/nl/binaries/20140403_AM_ reserve_strategique_electricite_tcm325-245763.pdf, accessed 1 February 2015. 78 Ministerial Decision of 16 July 2014, available at http://economie.fgov.be/nl/binaries/AMinstruction%20au%20gestionnaire%20r%C3%A9seau%20de%20contracter%20volume%20compl%C3% A9mentaire%20r%C3%A9serve%20strat%C3%A9gique_sign%C3%A9_tcm325-252938.pdf, accessed 1 February 2015. 79 Elia, Analyse van volume in het kader van strategische reserves—maart 2014, 20 March 2014. 80 Federal Energy Administration, Advies van de Algemene Directie Energie over de noodzakelijkheid om een strategische reserve aan te leggen voor de winterperioden 2014–2015, 2015–2016 en 2016–2017, 2 April 2014. 81 Elia, Procedure for constitution of strategic reserves. Applicable for the 2014 tendering, available at http://www.elia.be/~/media/files/Elia/About-Elia/Users%20Group/Task-force-balancing/Strategic% 20Reserves/UK_2014_Procedure_for_constitution_of_Strategic_Reserves.pdf, accessed 1 February 2015. CREG, Nota over het ontwerp van de proceduremodaliteiten voor de aanleg van strategische reserves, 24 April 2014, available at http://www.creg.info/pdf/Diversen/Z1327NL.pdf, accessed 1 February 2015. 82 Elia, Werkingsregels voor de strategische reserve, available at http://www.elia.be/nl/over-elia/ users-group/Strategic-Reserves-Implementation-Task-Force/Werkingsregels-voor-strategische-reserves, accessed 1 February 2015. CREG, Eindbeslissing over het voorstel van de NV Elia System Operator betreffende de werkingsregels van de strategische reserve, 5 June 2014, available at http://www.creg.info/ pdf/Beslissingen/B1330NL.pdf, accessed 1 February 2015. 77

13.4 European dimension

253

notification. This means that in case of mothballing, a minimum period of ten months must be respected, and in case of closure this minimum period is extended to fifteen months. Lastly, no mothballing or closure may occur during the winter period which runs from 1 November until 31 March.

13.4 European dimension 13.4.1 State tendering for new capacities The tender specifications published by the Federal Energy Administration on 15 January 2014 clearly label the subsidy payment paid out to the winning bid as state aid.83 It is thus expected that the Belgian government will seek compliance under the EU state aid rules enacted in Articles 107 and 108 TFEU. The Q&A relating to the tender procedure of 15 January 2014 state that the Commission has been informed about the envisaged aid system though it is unclear if a formal state aid notification was made by the Belgian State or not.84 The tender specifications itself actually mentions that the Federal Energy Administration ‘may withdraw the request for proposals in case the Commission qualifies the financial incentive as unlawful state aid’.85 There could indeed be a problem as to the appropriateness of the aid given the discriminatory nature of the tender. The tender first of all rules out existing plants.86 This discrimination between new and existing power plants is problematic under the new EEAG 2014–2020, as these demand that any incentives are provided to ‘both existing and future generators’.87 The nondiscrimination principle also applies with respect to the type of technology (principle of technology neutrality) whereas this is not respected either as the Federal Energy Administration stipulated that only OCGT and CCGT plants can participate.88 Equally problematic under EU law is that the tender is not open to interconnections nor to demand side participation.89 Whilst the Federal Energy Administration clarified that the tender allowed for generation based abroad (two of the three bids made actually come from plants located in the Netherlands), interconnection capacity was not an option.

83

Tender specifications (n 48) p 21. See Q&A related to the tender (n 50) questions 20–22. The EU state aid registry does not contain a notification from Belgium regarding the envisaged aid system (http://ec.europa.eu/competition/state_aid/ register, accessed 1 February 2015). 85 Tender specifications (n 48) para 14. 86 See Q&A related to the tender (n 50) question 67. 87 Communication from the Commission, Guidelines on State aid for environmental protection and energy 2014–2020 [2014] OJ C 200/1–55 (EEAG 2014–2020). Also see Commission staff working document, Generation Adequacy in the internal electricity market—guidance on public interventions, 5 November 2013 (Generation Adequacy SWD): ‘The intervention must be transparent and non-discriminatory’ (p 21) and ‘capacity mechanisms open to capacity retention as well as new investments, without discrimination between the two categories ensure cost-effectiveness and minimise distortion’ (p 27). 88 EEAG 2014–2020 (n 87) para 227. Also see Generation Adequacy SWD (n 87) p 26, which states that ‘it is more cost-effective and less distortionary to the internal market to base any restrictions on participation in the mechanism on performance specifications’ instead of a public authority establishing a specific technology. 89 EEAG 2014–2020 (n 87) para 233(a). 84

254

Belgium

Questions of lawfulness of the state tendering may arise not only under EU state aid rules but also under EU energy internal market rules. Article 8(4) of the 2009 Electricity Directive is precisely meant to ensure that increasing capacity from existing power plants as well as re-commissioning of moth-balled plants is also considered under the state tendering whilst these plants are excluded under the tender.90 The definition of ‘tendering procedure’ in the 2009 Electricity Directive also make clear that this includes existing capacity.91 Given that there is no experience to date with state tendering in Belgium, it will have to be seen whether or not the tender will ‘distort normal market operation or production decisions or to distort future investment decisions’,92 especially in view of the discrimination between types of technologies and between existing and new capacity. There could also be risks of possible ‘overcompensation’ in case the aid amount has not been well designed and given that it remains in place for a fixed duration of six years (though some sort of correction is provided in the tender specifications).93

13.4.2 Strategic reserve In its study dated 11 October 2012,94 the CREG discussed, compared, and analysed existing capacity mechanisms in Spain, Portugal, Ireland, Sweden, and Finland. It also looked at proposed capacity mechanisms in France, Germany, UK, Italy, Spain, and the Netherlands. In doing so it evaluated the mechanisms in terms of the results achieved and the regulatory and legal design. The Energy Regulator concluded that capacity mechanisms entail risks in terms of jeopardizing the internal energy market and that any such mechanism should be a measure of absolute last resort as well as temporarily. The Energy Regulator also emphasized the need for cooperation between neighbouring countries before implementing any (national) capacity mechanisms. In the draft bill, the government mentioned that the proposed mechanism of strategic reserve has been designed in liaison with other Member States as provided for in Art 3(2)(c) of the Security of Supply Directive.95 This consultation is—according to the government—also in line with the Commission’s recommendation as expressed

90 Christopher Jones, Floris Gräper, and Christof Schoser, ‘Security of supply’ in Christopher Jones (ed), The Internal Energy Market. The Third Liberalisation Package (Claeys & Casteels, 2010) p 545. Art 8(4) of the 2009 Electricity Directive (n 23) reads: ‘In invitations to tender for the requisite generating capacity, consideration must also be given to electricity supply offers with long-term guarantees from existing generating units, provided that additional requirements can be met in this way.’ 91 Art 2(24) of the 2009 Electricity Directive (n 23) reads: ‘the procedure through which planned additional requirements and replacement capacity are covered by supplies from new or existing generating capacity.’ 92 Generation Adequacy SWD (n 87) pp 22–3. See also the EEAG 2014–2020 (n 87) para 233. 93 Generation Adequacy SWD (n 87) p 35: ‘does the chosen mechanism deliver a price of zero when there is already sufficient capacity available?’ 94 CREG’s study on capacity mechanisms (n 1). 95 Draft bill (n 58) p 5. Directive 2005/89/EC of the European Parliament and of the Council of 18 January 2006 concerning measures to safeguard security of electricity supply and infrastructure investment [2006] OJ L 33/22 (Security of Supply Directive).

13.5 Conclusion

255

in the Generation Adequacy SWD.96 In the explanatory memorandum to the draft bill, the government continues by giving an overview of the different meetings held with relevant authorities, energy regulators, and/or transmission system operators from the Netherlands, Germany, France, Austria, Switzerland, and the UK.97 The government did not express a view regarding the compatibility of the scheme under EU state aid rules, though the Raad van State/Conseil d’Etat (an advisory and jurisdictional institution which must be consulted on draft legislation) considers that on the basis of PreussenElektra,98 there is no state aid as the compensation paid out to the participating power plants is covered through a public service tariff which is borne by the consumer.99 The exclusion of operational power plants as well as foreign power plants and interconnections from participating in the strategic reserve may raise questions under the discriminatory principle as discussed in section 13.4.1 and under the under the free movement rules.100 Similarly, the obligation for certain power plants to participate and make a bid as well as the possibility for the government to fix the capacity price could raise various legal issues under the free movement rules.

13.5 Conclusion It is indisputable that measures are needed to ensure future security of supply in Belgium in light of the nuclear phase-out and underinvestment. The Belgian authorities quickly designed two measures in 2014 (new capacity tendering and strategic reserve) with a rather aggressive implementation timetable. The design of these measures is legally flawed on a couple of points. More importantly, what is unclear at this stage is if and how the measures will transform the power market and what the commercial implications will be. Most likely, the design of the capacity mechanisms will evolve in the future and so will the law.

96 Generation Adequacy SWD (n 87) p 14: ‘Member States considering public interventions to ensure generation adequacy are invited to cooperate with Member States in their region at an early stage, to examine the potential of implementing cross border mechanisms.’ 97 Draft bill (n 58) p 6. 98 Case C-379/98 PreussenElektra AG v Schhleswag AG [2001] ECR I-2099. 99 Draft bill (n 58) p 30. 100 Free movement of goods (Arts 34–36 TFEU), services (Arts 56–57 TFEU) and capital (Art 63 TFEU). See chapter 11 for a discussion.

14 France Daniel Crevel-Sander and Charlotte Beaugonin1

14.1 Introduction This chapter offers a contribution to the on-going discussions at European level regarding generation adequacy based on the French experience. After a brief section describing the context in which the French authorities decided to implement a capacity mechanism to set the scene (section 14.2), we describe in detail the French capacity mechanism (section 14.3) and discuss legal implications under European sectorspecific law, antitrust rules and state aid rules (section 14.4).

14.2 Setting the scene The current situation in the French electricity sector is the result of political decisions made during the period from the 1950s to 1980s. These include, for example, the development of hydro and nuclear power2 as well as electric heating. While the former provide for a significant volume of base-load generation, the latter contributed to the creation of a phenomenon of consumption peaks particularly during cold periods. The majority of installed capacities (over 70 per cent of installed capacities) are still owned by EDF, the former legal monopoly. For the time being, the French electricity market is an energy-only market, but a capacity mechanism is currently being implemented, to be effective by 2016.

14.2.1 From the late 1990s to the Sido-Poignant Report in April 2010 Since the late 1990s, the consumption peak has been increasing at a faster rate than overall electricity consumption. Between 1997 and 2008, the difference between the average power demand in winter and the maximum power demand has increased by over a third, from 14 GW to 19 GW. Starting in 2005, electricity producers have been more and more concerned about the potential lack of investment in new capacities, and have informed the public authorities of the difficulties they face in recovering costs in the energy-only market.

1 The views and opinions expressed in this chapter are those of the authors and do not necessarily reflect the official policy or position of the EDF Group, to which the authors are currently affiliated. 2 Representing together around 70% of total installed capacities (128 GW) and accounting for over 85% of all electricity produced in 2012 (541 TWh).

14.2 Setting the scene

257

In 2012, a real risk of energy shortage in winter 2015–2016 has been predicted by RTE, the French TSO.3 Consequently, the French government took two separate steps to address the issue of security of electricity supply. As a first step, it adopted a new multi-year plan of investment (MPI). As a second step, it established a working group to study the consumption peak phenomenon and make proposals on how best to manage it.

14.2.1.1 Multi-year plan of investment (MPI) The MPI is used to define an objective of installed electricity generation capacity split by energy sources and, where appropriate, by technical means4 and geographic areas.5 The MPI for years 2009–2020 was first presented before the French parliament in the form of a governmental report, later published on 3 June 2009.6 It was subsequently adopted by two Ministerial Decrees on 15 December 2009.7 The report details the objectives of the energy policy (security of supply, environmental protection and competitiveness) in terms of development of generation capacity until 2020. As far as generation adequacy is concerned, it emphasizes that the proper functioning of the electricity system requires some thermal generation capacities.8 The report notes, however, that ‘in the opinion of all stakeholders, the market does not provide a return on investment for new capacity’.9

14.2.1.2 The working group The working group, co-chaired by a member of the French National Assembly, Serge Poignant, and a member of the French Senate, Bruno Sido, was created immediately after the publication of the MPI report. Its objective was to identify peak demand for electricity and study technical solutions to reduce it through demand response. Further, the group was asked to investigate ways to increase demand response and make proposals promoting demand response measures, rather than the construction of new power plants. In April 2010, the working group issued a report containing of twenty-two proposals (the Sido-Poignant Report). As far as generation adequacy is concerned, the report stresses two main market failures in the French market. First of all, it confirms the lack 3

RTE, Bilan prévisionnel de l’équilibre offre-demande d’électricité en France (2012 edn) p 10. Meaning the MPI may specify the preferred technical solution to generate electricity from a given source. 5 French Energy Code, Art L.141-1 ff. 6 Ministry of Ecology, Sustainable Development and Energy (Ministry for Energy), Report to the Parliament relating to the multiannual plan of investments for 2009–2020, June 2009, available at http:// www.developpement-durable.gouv.fr/IMG/pdf/ppi_elec_2009.pdf, accessed 1 February 2015 (Report to the Parliament). 7 Ministerial Decrees, French Official Journal, 10 January 2010, sections 2 and 3. For example, as far as renewable energies are concerned, the MPI provides the following development objectives: 25,000 MW for wind power (split among 19,000 MW onshore and 6,000 MW offshore); 5400 MW for solar energy; 2300 MW for biomass; 3 TWh per year and 3,000 MW of peak capacity for hydropower. 8 Report to the Parliament (n 6) p 13. 9 Report to the Parliament (n 6) p 14. 4

258

France

of investment in new capacity.10 Second of all, it highlights the absence of mechanisms to develop demand response. With respect to the latter, the report notes that enhancing demand response participation in the energy-only market is ‘doomed to failure because even if energy-only markets can theoretically be profitable for peak generation capacity (and symmetrically for demand response measures) the visibility they offer is not sufficient. Price spikes are too random in frequency and level and the risk is too great for an investor. ( . . . ) No supplier has an interest to assume the risk of such an investment, to the extent that any failure would not necessarily be his responsibility and does not necessarily lead to unbearable losses.’11 In view of these concerns and security of supply measures taken in other countries,12 the report includes a proposal to establish a capacity mechanism in France consisting of a capacity obligation of suppliers, along with a certificate market.13 At the same time, the report highlighted the need for considerable additional work to implement these proposals and stated that in order to be effective, by 2015, the capacity mechanism should be implemented by the end of 2011, which did not happen.

14.3 Capacity mechanism In April 2010, the government tabled a bill before the French parliament on the New Organisation of the Electricity Market (Nouvelle Organisation du Marché de l’Électricité, the NOME Law).14 Adopted on 7 December 2010, the NOME Law reflects the proposals of the Sido-Poignant Report. Article 6 of the NOME Law defines the key principles of a capacity mechanism, to be implemented by a governmental decree. These stipulations have since been codified in 2011 in Articles L.335-1 to L.335-6 of the French Energy Code15 and are discussed in more detail in section 14.3.1 below. The decree implementing the capacity mechanism was adopted on 14 December 2012 (the Capacity Decree, discussed in section 14.3.2 below).16 According to this decree, the French Minister of Ecology, Sustainable Development and Energy (energy minister) shall adopt technical rules for the capacity mechanism. A proposal shall be prepared by the TSO, following an opinion of the Commission de Régulation de l’Energie (CRE), the energy regulator.17 The French capacity mechanism was expected 10 The Sido-Poignant Report, April 2010, available at http://www.developpement-durable.gouv.fr/IMG/ pdf/Rapport_Poignant-Sido.pdf, accessed 1 February 2015, p 11: ‘Most generation facilities are ageing. ( . . . ) the growth of consumption peaks and obsolete plants call for measures to ensure that investment in both demand response and generation is made. Beyond renewable energy, investment in semi-base and peak capacity is also an issue.’ 11 The Sido-Poignant Report (n 10) p 20. 12 The Sido-Poignant Report (n 10) p 20. 13 The Sido-Poignant Report (n 10) proposals n 16 and 17. 14 Law 2010–1488 of 7 December 2010 (Loi n 2010–1488 du 7 décembre 2010 portant nouvelle organisation du marché de l’électricité), (the NOME Law). 15 Law 2013–312 of 15 April 2013 (Loi n 2013–312 du 15 avril 2013 visant à préparer la transition vers un système énergétique sobre et portant diverses dispositions sur la tarification de l’eau et sur les éoliennes). 16 Decree 2012–1405 of 14 December 2012 (the Capacity Decree). An association of alternative electricity suppliers has introduced an action for annulment of the Capacity Decree on 17 June 2013 in front of the French Council of State. A decision is unlikely to be handed down before end of 2014. 17 See section 14.3.3 for details.

14.3 Capacity mechanism

259

to become operational as early as 2014 with the first certification of generation or demand response capacity for the winter of 2016–2017.

14.3.1 The NOME Law 14.3.1.1 The parliamentary debate on the NOME Law The parliamentary debates on the NOME Law offer interesting insights into the objectives pursued by the French legislator regarding the creation of a capacity mechanism. The energy minister defended the provisions of the NOME Law before the National Assembly,18 stressing that a capacity mechanism would ‘strengthen the security of supply in France by requiring all suppliers to have, directly or indirectly, enough generation capacity or demand response measures to supply their customers at all times’.19 He also highlighted another key objective of the legislator, namely that the capacity mechanism would ‘ensure that all suppliers assume all their industrial and energy ( . . . ) responsibilities regarding their customers and do not rely on an implied security of supply provided by the electricity incumbent’.20 The implied security of supply refers to the regulated wholesale access also created by the NOME Law, known as ARENH.21 This mechanism allows alternative electricity suppliers to purchase nuclear energy from EDF, the incumbent supplier, on a transitional basis until 2025 and under regulated conditions. This system was designed to meet the concerns expressed by the Commission about the insufficient competition within the French electricity market.22 In the context of ARENH, it was underlined during the debates before the Senate that the capacity mechanism was designed to offset the right of alternative electricity suppliers to access nuclear power, as it would ‘oblige them to contribute to the [country’s] security of supply.’23 Moreover, when designing the capacity mechanism, the French government relied on foreign experiences with capacity mechanisms in addition to analysing possible economic effects of implementing a capacity mechanism in France. When presenting the NOME Law before the National Assembly, the energy minister specifically referred to the United States.24 In particular during the parliamentary process, the government often highlighted that the obligations imposed upon electricity suppliers would not result in additional costs. A report on the economic impact of the NOME bill was submitted to the parliament while the NOME bill was being examined. According to this impact assessment, an efficient capacity mechanism should ensure, on the one hand, that a

18

The National Assembly is the lower house of the bicameral Parliament of France. The upper house is the Senate. 19 Preamble of the NOME Law, available at http://www.assemblee-nationale.fr/13/pdf/projets/pl2451. pdf, accessed 1 February 2015, p 9. 20 Preamble of the NOME Law (n 19) p 9. 21 ARENH stands for Regulated Access to Incumbent Nuclear Electricity. 22 For more information on ARENH see RTE’s website at https://clients.rte-france.com/lang/an/clients_ producteurs/services_clients/dispositif_arenh.jsp, accessed 1 February 2015. 23 Mr Poniatowski, rapporteur of the bill and member of the Economy Commission of the Senate, 29 September 2010. 24 Preamble of the NOME Law (n 19) n 9.

260

France

single price of capacity is determined and, on the other hand, that market players can exchange capacity certificates with each other. In such cases, the report notes that ‘all stakeholders (producers, suppliers, consumers) would be encouraged to provide capacity or demand response and the implemented solutions will be the most relevant in economic terms’.25 As regards the price of the capacity certificate, the report first stresses that this price cannot be anticipated, given the various factors that may affect its level each year. However, the document notes that ‘the price may not exceed the fixed cost of generation means, and that new plants would be built until the supply capacity meets demand. As a first approximation, this analysis assumed that the price of the capacity certificate is therefore capped at a fixed cost of a combustion turbine. This fixed cost here is estimated at €60,000 per MW per year’.26 The capacity mechanism is designed so as to ensure that the price of capacity certificates will tend towards zero if there are no generation adequacy concerns (or overcapacities). Otherwise, the price of capacity certificates will reflect the value of a failure of the electricity system.

14.3.1.2 The design of the French capacity mechanism provided by the NOME Law The French capacity mechanism is based on new obligations imposed by the NOME Law on both electricity suppliers and operators of generation and/or demand response capacities. Electricity suppliers are required to contribute to the security of supply in proportion to the electricity consumption of their customers.27 They are therefore required by law to hold, for each given period, a specific amount of capacity certificates determined by the TSO (see section 14.3.2.3 below for further details), demonstrating that they are able to cover the needs of their customer portfolio in the periods during which the consumption of all customers is the highest.28 All operators of generation capacity and/or demand response are required by law to contribute to the security of supply by certifying all of their capacity through a contract to be concluded with the TSO. The contract, the model of which does not exist yet, will determine the level of capacity to which the operator commits to make available in the future, the verification modalities and the sanctions in case the operator does not fulfil its capacity commitment.29 Capacity certificates are exchangeable and transferrable.30 Capacity certificates offered by operators of generation and/or demand response capacity are matched with bids by electricity suppliers (willing to purchase these certificates) via a marketbased mechanism (certificate market). The resulting price of capacity certificates reflects the cost of the capacity necessary to ensure the safety of the system.

25 For an impact assessment of the NOME Law, go to http://www.assemblee-nationale.fr/13/pdf/projets/ pl2451-ei.pdf, accessed 1 February 2015 (impact assessment). 26 Impact assessment of the NOME Law (n 25) pp 26–7. 27 28 French Energy Code, Art L.335-1 para 1. French Energy Code, Art L.335-2 para 1. 29 30 French Energy Code, Art L.335-3 paras 1–2. French Energy Code, Art L.335-3 para 2.

14.3 Capacity mechanism

261

14.3.1.3 Principles of the French capacity mechanism defined by the NOME Law As intended by the French legislator, the system should operate with a sufficient anticipation in order to leave enough time for investors to develop new generation capacity or demand response measures to fill a potential adequacy gap threatening the security of supply.31 The French capacity mechanism is based on the multiannual generation adequacy forecast required by law and prepared by the TSO.32 It applies to the continental French metropolitan territory33 and ‘takes into account the interconnection of the French market with other European markets’.34 The impact of this principle is further detailed in the Capacity Decree (see below, section 14.3.2). For electricity suppliers and big customers, the NOME Law imposes the obligation to contribute to the security of supply, with specificities for (a) local distribution companies,35 (b) customers who obtain part of their electricity needs directly on wholesale markets, and (c) network operators for their network losses.36 The law also provides for a specific regime for the electro-intensive consortium known as Exeltium37 and for electricity sold by EDF to alternative suppliers under the ARENH mechanism (see section 14.3.1 above).38 For generators and/or demand response providers, the aim of the French government is to implement a technology-neutral system (ie not discriminating between different types of generation capacities as well as between supply side and demand side measures).39 However, considering the political will to favour the development of demand response as an alternative to electricity generation, the NOME Law grants a priority, at equal cost, to demand response measures over generation capacities.40 The certification methods are to take into account technical specificities of various types of generation capacity or demand response measures, be transparent and non-discriminatory.41

14.3.2 The Capacity Decree The Capacity Decree details the implementation of the French capacity mechanism in technical terms. The following sections only focus on certain relevant key legal aspects 31

32 French Energy Code, Art L.335-2 para 4. French Energy Code, Art L.141-1. French Energy Code, Art L.335-1 para 1. No subdivision of the national territory is anticipated. 34 French Energy Code, Art L.335-2 para 3. 35 Such companies hold local distribution and regulated tariffs supply monopolies. 36 French Energy Code, Art L.335-5 para 2. See also Capacity Decree (n 16) Art 21 II and III. 37 French Energy Code, Art L.335-5 para 3. 38 Capacity Decree (n 16) Art 21 I. See also Decree 2011–466 of 28 April 2011 (Décret n 2011–466 du 28 avril 2011 fixant les modalités d’accès régulé à l’électricité nucléaire historique) Art 1 V. 39 Energy Minister, ‘Une obligation de capacité pour réduire la pointe électrique et garantir la sécurité d’approvisionnement’, Dossier de presse, 19 December 2012, available at http://www.developpementdurable.gouv.fr/Une-obligation-de-capacite-pour.html, accessed 1 February 2015. 40 French Energy Code, Art L.335-2 para 3 in fine. Emphasis added. 41 French Energy Code, Art L.335-4 para 1. 33

262

France

for the comparison of various capacity mechanisms in Europe and their analysis under the European law.42

14.3.2.1 Determination and control of the capacity obligation of each electricity supplier The determination of the capacity obligation imposed upon each electricity supplier delivering electricity to a customer based on the continental French metropolitan territory follows a three step process.43 First, the TSO determines, four years in advance, the overall capacity requirement to match the total French electricity demand for a given year N during the periods of highest consumption. Second, this overall capacity requirement is reduced by taking into account the contribution of interconnections based on statistical import flows. Finally, this overall capacity requirement is spread among all electricity suppliers taking into account the forecasted electricity demand of their customers during consumption peaks to be further defined by technical rules. In year N-4, the TSO issues a forecast about the capacity required in year N enabling electricity suppliers to estimate their own capacity obligation for that year. From year N-4 until year N, capacity certificates may be purchased and/or sold (see section 14.3.2.3 below). For each year N, the TSO sets a deadline for capacity certificate transactions and notifies electricity suppliers of the exact amount of their capacity obligation no later than fifteen days before this deadline. Immediately after the deadline for capacity certificate transactions, the TSO calculates whether electricity suppliers hold enough capacity certificates to cover their capacity obligation. In order to fulfil their capacity obligation, electricity suppliers can either purchase capacity certificates from the existing operators of generation and/or demand response capacity or, alternatively, they themselves can invest in electricity generation and/or demand response measures. The obligations of electricity suppliers are sanctioned by a bonus-malus system. Those who exceed their obligation (ie hold more certificates than required) are entitled to a bonus payment financed by the malus payments of electricity suppliers who do not match their obligation (ie hold fewer certificates than required). It must be noted that the total amount of bonus payments cannot exceed the total amount of malus payments, and the TSO will set up a separate account for settling these payments.44

42 In particular, the following developments do not discuss specific provisions of the Capacity Decree which provide for the possibility for the energy regulator, during the six first years of the ramp up of the capacity mechanism, to launch tenders for the construction of new electricity generation capacities in the case of an ‘exceptional risk of inadequacy between offer and demand.’ See Capacity Decree (n 16) Arts 22–23. Art 26 also provides for a specific tender to be launched by the CRE in order to ensure security of supply for the winter 2015–2016 for which it is anticipated that the capacity mechanism will not yet be sufficient to fully play its role. 43 Capacity Decree (n 16) Arts 3–4. 44 Capacity Decree (n 16) Arts 5–6.

14.3 Capacity mechanism

263

14.3.2.2 Certification and control of electricity generation or demand response capacities Each operator of generation and/or demand response capacity is obliged, four years ahead of each delivery year, to file a request for certification of its generation capacity and/or demand response measures with the TSO or, alternatively, with the relevant DSO depending on which network its capacity is (to be) connected to. This obligation applies both for existing45 and planned capacities, while the latter require more detailed specifications by means of technical rules.46 Furthermore, each operator needs to conclude a contract with the TSO specifying its capacity commitment, modalities of correction of the planned availability, control modalities and penalties in case the committed availability is not reached.47 After the peak period of a year N, the TSO controls whether the level of availability committed to by each participant to the capacity mechanism has been reached. The commitments of capacity operators are also sanctioned by a bonus-malus system. Those who exceed their capacity commitment (ie actual available capacity higher than planned) are entitled to a bonus payment financed by the malus payments of capacity operators who do not match their obligation (ie actual available capacity is lower than planned). Similarly, in this case the total amount of bonus payments cannot exceed the total amount of malus payments, and the TSO will run a separate account set up for settling these payments.48

14.3.2.3 Organization of capacity certificates exchanges and transactions Capacity certificates are to be traded through a capacity register administered by the TSO, where all electricity suppliers and capacity operators hold capacity accounts.49 The financial payment resulting from a given capacity transaction will be directly and privately settled between the parties to the transaction. Any transfer of capacity certificates has to be declared to the TSO jointly by both parties to the transaction.50 Any commercial modalities of the transaction (including the price) have to be communicated to the energy regulator which publishes yearly statistics on transaction volumes and prices.51 At a later stage, and based on the energy regulator’s yearly reports on the functioning of the capacity mechanism, the energy minister has the possibility of creating an organized market for capacity certificates.52 Finally, the energy regulator is required to issue an annual report on the interactions of the French capacity mechanism with neighbouring markets as well as its integration within the European market and, where necessary, to suggest improvements to its design. 45 The specificity of small capacities is taken into account in order to ensure that the costs of certification and control remain significantly lower than their contribution to the security of supply, which is to be achieved through aggregation of such capacities. See Capacity Decree (n 16) Art 10. 46 Capacity Decree (n 16) Art 8 I of the Capacity Decree. 47 Capacity Decree (n 16) Arts 8 II, 9, and 11. Concrete modalities are not determined yet. 48 49 Capacity Decree (n 16) Art 14. Capacity Decree (n 16) Arts 15 and 16 I. 50 51 Capacity Decree (n 16) Art 16 IV. Capacity Decree (n 16) Art 17 I and II. 52 Capacity Decree (n 16) Art 19.

264

France

14.3.3 Status of the adoption of technical rules Since the publication of the Capacity Decree, the TSO has been consulting extensively with all stakeholders on the development of technical rules.53 Their adoption was first planned before 1 November 2013, however, due to their complexity, the process has been delayed. The TSO eventually proposed a set of technical rules to the energy minister on 9 April 2014,54 on which the energy regulator gave its opinion on 28 May 2014.55 The energy minister has not approved this proposal yet.

14.4 European dimension One of the recurring issues in the debate on the French capacity mechanism has been its compatibility with EU law, in particular with respect to the internal energy market directives and free movement rules, competition rules, and state aid rules.

14.4.1 Compatibility with the internal energy market directives and free movement rules Opponents to the mechanism have, inter alia, claimed that it would violate the principle of free movement of goods56 enshrined in the Treaty, but that it would also breach both the Security of Supply Directive57 and the 2009 Electricity Directive.58 In this respect, it was mainly highlighted that the mechanism only applies to generation and/or demand response capacity located within the French metropolitan territory, thus effectively excluding foreign capacity from participating in the mechanism. On the supplier side, any electricity supplier active in France is under an obligation to demonstrate that it holds a sufficient number of capacity certificates corresponding to capacities located within the French metropolitan territory. In particular, it was claimed during discussions amongst stakeholders that the scheme could be considered a restriction of free movement of goods safeguarded by Articles 34 and 35 TFEU and/or that it would create new barriers to entry into the French market thus jeopardizing the completion of the EU internal energy market in violation of the previously mentioned directives. 53

See the introduction of section 14.3. RTE, Mécanisme de capacité—proposition de règles et dispositions complémentaires, 9 April 2014, available at http://www.rte-france.com/uploads/Mediatheque_docs/vie_systeme/annuelles/Mecanisme_ capacite/20140409-Regles-Mecanisme-Capacite.pdf, accessed 1 February 2015. 55 CRE, Opinion of 28 May 2014, available at http://www.cre.fr/documents/deliberations/avis/ regles-du-mecanisme-d-obligation-de-capacite/regles-du-mecanisme-d-obligation-de-capacite, accessed 1 February 2015. 56 For an in-depth discussion on capacity mechanisms and the free movement of goods, see chapter 11. 57 Directive 2005/89/EC of the European Parliament and of the Council of 18 January 2006 concerning measures to safeguard security of electricity supply and infrastructure investment [2006] OJ L 33/22 (Security of Supply Directive). 58 Directive 2009/72/EC of the European Parliament and of the Council of 13 July 2009 concerning common rules for the internal market in electricity and repealing Directive 2003/54/EC [2009] OJ L 211/55 (2009 Electricity Directive). 54

14.4 European dimension

265

According to the Security of Supply Directive, security of supply could be more efficiently addressed at the European level than at a national level, given the scale and effect of the security of supply measures.59 Accordingly, the French competition authority (competition authority) noted in its opinion on the French capacity mechanism60 that capacity mechanisms should be adopted at the EU level, drawing an analogy to the EU ETS.61 However, it seems that the idea of a pan-European capacity mechanism is far-fetched as it would raise significant practical problems given the country-specific nature of the security of supply issues. Furthermore, the security of supply directive does not suggest any particular measures at the EU level, but instead offers the possibility of implementing national mechanisms. Some representatives of the French government express the view that the French capacity mechanism is fully compatible with EU market design.62 The proponents of this view rely on possible justifications or derogations under various EU provisions which might be of relevance for this assessment. These are (a) Article 36 TFEU, (b) the Security of Supply Directive, and (c) PSOs. First of all, capacity mechanisms are usually implemented for security of supply reasons. The French government openly states that the French capacity mechanism has been introduced in order to secure energy supplies.63 The objective of security of energy supply has been recognized by the CJEU as one of the possible justification grounds listed in Article 36 TFEU for a restriction of the free movement of goods.64 Second of all, the Security of Supply Directive explicitly allows member states to adopt national measures to ensure generation adequacy provided that they are nondiscriminatory and market based.65 Lastly, the 2009 Electricity Directive allows Member States to impose PSOs also for security of supply reasons, provided that they are ‘clearly defined, transparent, non-discriminatory, verifiable and shall guarantee equality of access for electricity undertakings of the Community to national consumers’.66

59

Security of Supply Directive (n 57) Recital 18. Before the adoption of the Capacity Decree, the French Government asked the competition authority for its opinion on the project which it rendered 12 April 2012. See Autorité de la Concurrence, Opinion n 12-A-09 of 12 April 2012. 61 Autorité de la Concurrence (n 60) para 132. 62 Antoine Pellion, ‘Necessity, design and implementation of decentralised capacity obligations in France’, presented on behalf of the General Directorate for Energy and Climate European (Service Electricity Generation) at the European Workshop on capacity mechanisms in EU power markets: Are they necessary? How can we harmonize them?, 16 April 2013, University Paris-Dauphine, available at http://www.ceem-dauphine.org/agenda/fr/e850958bc345bba0229a08fb4c3c173ec1e63aab, accessed 1 February 2015, slides 10, 12, 14, and 15. 63 See the press file prepared for the launch of the French capacity mechanism by the then French energy minister, Delphine Batho (n 39). 64 Case C-483/99 Commission v France [2002] ECR I-04781 and Case C-207/07 Commission v Spain [2008] ECR I-00111. 65 See Security of Supply Directive (n 57) Recital 10 and Art 5(2). According to Recital 10, ‘[m]easures which may be used to ensure that appropriate levels of generation reserve capacity are maintained should be market-based and non-discriminatory and could include measures such as contractual guarantees and arrangements, capacity options, or capacity obligations. These measures could also be supplemented by other non-discriminatory instruments such as capacity payments.’ 66 See 2009 Electricity Directive (n 58) Art 3(2). 60

266

France

The co-existence of these three (at least) different legal provisions67 which could potentially be applied to capacity mechanisms raises questions of a coherent interpretation of EU law. In any case, the legal analysis under any of these provisions centres around the question of whether or not the French capacity mechanism is non-discriminatory, transparent, market-based and the most proportionate measure to achieve the objective of security of supply constituting at the same time the least possible restriction to cross-border trade. However, testing the capacity mechanism before its actual implementation may prove to be a very difficult exercise in practice. With respect to the criterion of being transparent and market based, it has thus far never been disputed that the French capacity mechanism as described in section 14.3 is in fact market based and provides for a great deal of transparency. With respect to the issue of non-discrimination, two different aspects have been raised. A possible discrimination amongst the French market players (which is not a concern of free movement rules but possibly of competition law68) as well as a possible discrimination between market players active in France and those active in other European countries. In this latter respect, it should first be noted that foreign capacities cannot participate in the French capacity mechanism due to technical reasons, as the competition authority acknowledged in its opinion.69 Under the current market conditions, interconnectors to France constitute bottlenecks and their use is, in most cases, not available to market players. Instead they are used by transmission system operators and power exchanges in the context of market coupling. In other words, operators of capacity located abroad are neither technically nor legally in a position to make contractual commitments to contribute to France’s security of supply, particularly in the case of system stress. Nevertheless, the NOME Law provides that the contribution of interconnectors to the French security of supply is to be taken into account. Initially, it was thought to take interconnectors into account on a statistical basis, through a reduction of the overall need of generation capacity determined by the TSO before attributing specific capacity needs to each electricity supplier. However, the competition authority noted that such an approach would de facto favour the dominant player which would, in proportion, benefit from the biggest reduction of its obligation thanks to the interconnectors.70 Several ways of correcting this bias are currently under discussion, one of which is to certify interconnectors as such, by attributing capacity certificates to the TSO.71 This would increase the number of capacity certificates on the market in proportion to the contribution of the interconnectors, but would also require legislative amendments. 67 Even four if considering state aid rules, but the latter would in any case apply to capacity mechanisms in addition to any of the other provisions of the Treaty or sector-specific secondary legislation. State aid issues are addressed below in section 14.4.3. 68 Eg see Case C-212/06 Gouvernement de la Communauté française et Gouvernement wallon [2008] ECR I-01683, para 33, or Case C-127/08 Metock e.a. [2008] ECR I-06241, para 77. Competition law aspects are addressed below in section 14.4.2. 69 Autorité de la Concurrence (n 60) paras 134–5. 70 Autorité de la Concurrence (n 60) paras 141–2. 71 Directly inspired by a proposal made by the competition authority. Autorité de la Concurrence (n 60) paras 143–4.

14.4 European dimension

267

Given the complexity and the lack of a substantial and reliable track-record of experience with capacity mechanisms, the balancing test should draw on the existing experience in other countries, at the same time leaving a significant margin of assessment to the French authorities to adapt it to national specificities. Finally, while this option has not been in debate in France, introducing reciprocity clauses into national capacity mechanisms would further benefit energy market integration in Europe and could make the issue of the compatibility of national capacity mechanisms with European rules easier to deal with. Such clauses would provide a possibility to exchange national capacity certificates against foreign certificates. However, implementing this system would require some coordination at the European level in order to ensure minimal compatibility between national capacity mechanisms.

14.4.2 Antitrust issues For the purpose of issuing an opinion on the proposed capacity mechanism, the competition authority carried out an overall competitive analysis of the French electricity market and expressed a series of concerns.72 First of all, the competition authority raised doubts as to the necessity of the proposed capacity mechanism noting that there would be no consensus amongst stakeholders on the need for such a mechanism. Additionally, the competition authority states that the government has not considered alternative measures to ensure security of supply, including strategic reserves and demand side response. More specifically, the competition authority was concerned that the introduction of a capacity mechanism could hamper competition on the electricity retail markets because of the additional complexity for alternative electricity suppliers73 and could even prompt some of them to reduce their activity or even exit the market.74 Further, the competition authority also pointed out a flaw in regulation in the French market due to the fact that the cost of capacity certificates will be fully reflected in the cost structure of regulated tariffs for end customers75 only starting from 2016 while alternative suppliers would need to start including the costs of capacity certificates into their own retail prices earlier. This situation could lead to a potential margin squeeze between regulated tariffs and free market prices.76 Lastly, the competition authority also highlighted the likely dominant position of EDF both as a capacity operator as well as electricity supplier purchasing capacities, given its current strong position in the generation and supply business. More specifically, the competition authority identified a risk of market manipulation and/or of margin squeeze as EDF might favour its retail branch when transferring capacity certificates from its upstream branches to downstream suppliers.77 72

For a discussion on antitrust issues see chapter 10. 74 Autorité de la Concurrence (n 60) para 45. Autorité de la Concurrence (n 60) para 73. 75 Until 31 December 2015 all electricity consumers still have the opportunity to choose state determined regulated tariffs. After this date, only households and small professionals will retain the possibility to opt for regulated tariffs. 76 Autorité de la Concurrence (n 60) paras 79–80. 77 Autorité de la Concurrence (n 60) paras 103–7. 73

268

France

The issue of capacity certificate transactions between upstream and downstream branches of the same entity (intercompany transactions) has been directly addressed by the Capacity Decree since a single legal entity being both a capacity operator and an electricity supplier will be under the obligation to open two different accounts in the capacity register run by the TSO. Thus, all capacity certificate transactions, including intercompany transactions, will be registered and accessible to the energy regulator and the competition authority enabling them to control the behaviour of market players. The overall market design for capacity certificate transactions has been (and still is) much debated amongst stakeholders. Even though the idea had been floated to have only a compulsory organized market, there will be only over-the-counter (OTC) transactions at the start-up of the French capacity mechanism. However, the NOME Law provides for a possibility to have an organized certificate market in the future (see section 14.3.1.3). Another issue which still remains open is whether all capacity operators will be obliged to offer all of their capacity certificates for sale, or whether those market players who are both capacity operators and electricity suppliers will have the possibility to transfer the capacity certificates through intercompany transactions.78 A compulsory market would solve any capacity retention issues whereas a free market enabling intercompany downstream transactions could potentially give rise to capacity retention practices. However, the fact that all capacity certificate transactions will in any case be tracked on the TSO’s capacity register makes it easier for regulatory and/or competition authorities to detect potential abusive practices. Finally, in its opinion the competition authority also raised concerns regarding a provision of the Capacity Decree granting the energy regulator powers to launch tenders for the construction of new capacities in the case of an ‘exceptional risk of inadequacy between supply and demand’. Such new capacities would then benefit from public support. According to the competition authority, such a possibility would go against one of the key principles of the French capacity mechanism, ie the individual responsibility of each electricity supplier. The competition authority fears that this possibility could invite opportunistic behaviour of market players by delaying investment decisions just to be considered eligible for tender and public support. In this context the competition authority emphasized that such capacity retention practices may be considered an abuse of a dominant position or concerted practices in cases where they involve several independent market players.79

14.4.3 Compatibility with state aid rules Some opponents of the French capacity mechanism claim that the mechanism in its current design is likely to infringe upon the EU state aid rules80 and its implementation

78 Being recalled that such transaction would in any case be traced on the capacity register which will be held by the TSO. 79 Autorité de la Concurrence (n 60) paras 121–8. 80 For a discussion on state aid issues, see chapter 9.

14.4 European dimension

269

requires a notification to the Commission.81 The French government on the contrary takes the view that this would not be the case, but mentions that discussions are to be held with the Commission in this respect.82 The French capacity mechanism has been compared to other systems which have been challenged under state aid rules. First, parallels have been drawn between Member States’ free allocation of emission rights to selected electricity generators under the EU ETS which, alternatively, have been sold to the benefit of national budgets.83 Secondly, the mechanism was also compared to the green certificate systems, which constrain operators to pay a penalty in case of non-compliance with the state imposed obligations.84 However, the French capacity mechanism differs from these systems on key design points, which may disqualify at least two of the four cumulative criteria required to qualify as state aid under Article 108 TFEU, namely (a) economic advantage and (b) intervention by the state or by means of state resources.85 Regarding (a), both the creation of an additional revenue stream for capacity providers and the bonus-malus system have been challenged on the grounds of providing an economic advantage. In fact, both cases are not clear-cut. Since the objective of the French capacity mechanism is to reach generation adequacy, the price of capacity certificates will tend towards zero as soon as the required capacity level is reached (see above, section 14.3.1.1). In this case, therefore, capacity certificates will not generate any additional revenues. Additional revenues, if any, will only occur in case of an adequacy gap, and will compensate the generators assuming capacity obligations at a market price reflecting the risk of a failure of the electricity system. Finally, the capacity certificates are not created ex nihilo but are attributed to capacity operators to reflect the availability of existing generation capacities thus only recognizing realized investments. The bonus-malus system is inherent to the French capacity mechanism and conceived so as to avoid both overinvestments and free-riding behaviour. Market players who decide to spend less money than necessary to meet their obligations and thus put the stability of the electricity system at risk will be penalized. Market players who, however, decide to spend more money than they are obliged to, will get a bonus for contributing to the system stability. But if no market player pays penalties, no market players will get a bonus even if they exceed their obligations, thus guaranteeing that the system only promotes virtuous behaviour.86

81 Arguments of opponents to the capacity mechanism have mainly be mentioned orally in various fora in which the French capacity mechanism was discussed. They are also reflected in the pending appeal against the Capacity Decree (n 16). 82 See Pellion (n 62) slide 12. 83 Case C-279/08 P Commission v Netherlands [2011] ECR I-07671, paras 106–8. 84 Case SA.33134 Green certificates for promoting electricity from renewable sources (Photovoltaics Romania) [2011] OJ C 244, paras 53–4. 85 The two other criteria, selectivity of the aid and impact on trade between Member States are less debated. 86 This bonus-malus system is almost identical to the imbalance settlement mechanism in place in France and in several European countries to guarantee intraday balancing of the electric system. To the authors’ knowledge, such mechanisms have never been challenged under state aid rules.

270

France

Regarding (b), the French capacity mechanism has no impact on the state budget, even in the form of a ‘sufficiently concrete economic risk of burdens’.87 Contrary to the other schemes quoted earlier as examples of state aid, the French State neither renounces to sell pre-existing state assets nor to forego eventual penalties. The French capacity mechanism merely organizes the transfer of private resources amongst capacity operators and electricity suppliers in order to ensure a sufficient level of investment generation capacities. Contrary to the capacity certificate transactions, which will be settled through registered accounts in the capacity register, any cash flows resulting from those privately negotiated transactions will be directly transferred between the parties. Only the bonus-malus payments will be settled through separate accounts registered with the TSO88 (see sections 14.3.2.1 and 14.3.2.2). Overall, if the generation adequacy gap raises the price of capacity certificates, the French capacity mechanism will in fact increase state revenues through the taxation of the additional revenues of capacity operators. Lastly, should the French capacity mechanism be considered by the Commission as involving state aid, its compatibility with the internal market would require a proportionality test, as discussed above in section 14.4.1.

14.5 Conclusion The French experience shows that the implementation of a capacity mechanism is a long-term endeavour raising several technical and legal issues to be dealt with. It clearly appears that it is, for the time being, impossible to envisage a unique European-wide response to the generation adequacy since a well-designed capacity mechanism requires taking into account national specificities in order to correctly address each specific generation adequacy issue. Nevertheless, some level of European coordination is desirable in order to prevent the development of national capacity mechanisms from jeopardizing the continuous integration of the internal energy market.

87

See Joint Cases C-399/10 P and C-401/10 P Bouygues and Bouygues Telecom v Commission (judgment of 19 March 2013, nyr) paras 109–110. 88 A 100% subsidiary of EDF itself is controlled by the French State.

15 Germany Kai Uwe Pritzsche and Katharina Reinhardt

15.1 Introduction Germany’s energy transition (Energiewende) faces major challenges. The current liberalized energy-only market, in which power plant operators are mainly paid for the amount of energy they supply, is transitioning to a system in which renewables are to become the central pillar of energy supply. However, electricity from renewables is regionally distributed differently from conventional electricity generation and its generation is volatile. For that reason, in some regions like the Northern German coast and East Germany there is an oversupply of electricity whereas in Southern Germany there is potential shortage after shutdown of nuclear power plants and the decision of phasing out the operation of the remaining ones between 2016 and 2022. In order to counter these imbalances, in the past Germany has introduced different kinds of strategic reserves to protect against regional and seasonal shortages, called a ‘network reserve’, which are meant to ensure the security of supply especially during the winter. After the September 2013 elections an in depth assessment was started in order to see whether other or further measures such as the introduction of a capacity mechanism are necessary. In November 2014, the Federal Ministry of Economic Affairs and Energy (BMWi) published a green paper on the Electricity Market for the Energy Transition1 in order to discuss whether the establishment of capacity mechanisms is necessary in order to guarantee the further functioning of the electricity market or whether a modernization of the energy-only market into an ‘Electricity Market 2.0’ will be sufficient.

15.2 Setting the scene 15.2.1 The challenge The German electricity market has to cope with the Energiewende as well as the nuclear phase-out. The energy transition was initiated with the adoption of the Energy Concept2 in September 2010 by the Federal Government. It sets out ambitious targets concerning the energy market and the reduction of greenhouse gases until 2050. It lays down measures for the development of renewable energy sources as well as network

1 BMWi, Ein Strommarkt für die Energiewende—Diskussionspapier des Bundesministeriums für Wirtschaft und Energie (Grünbuch) (November 2014) p 5. 2 BMWi, Energy Concept for an Environmentally Sound, Reliable and Affordable Energy Supply (2010/2011) (Energy Concept).

272

Germany

infrastructure. The Energy Concept3 tries to achieve two main objectives. First, it aims at a 40 per cent reduction in greenhouse gas emissions by 2020, and an 80–95 per cent reduction by 2050 with 1990 being the base year. Secondly, according to the Energy Concept, RES generation will account for 18 per cent of gross final energy consumption by 2020, and for 60 per cent by 2050. Due to the nuclear meltdown at Fukushima in March 2011, the role assigned to nuclear power in the Energy Concept was reassessed. The Federal Government decided to permanently shut down eight nuclear power plants right away and to phase out operation of the remaining nine nuclear power plants in Germany by 2022. On 6 June 2011 the Federal Government adopted the Energy Package4 that supplements the measures of the Energy Concept. These major changes have led to several problems. First, there is a general lack of investment in conventional generation capacity. Second, there is an imbalance in generation capacity with high generation capacity in Northern Germany and a severe lack in the Southern parts of Germany. Third, the German renewables support scheme was getting very expensive and it is attacked by EU state aid proceedings so that a fundamental reform of the Renewables Energy Act (EEG 2012)5 was urgently needed. The lack of investment in generation capacity is attributed to a number of factors. According to BMWi ‘the increasing sale of electricity from systems with low to very low marginal costs (particularly renewable energy systems)’ as well as ‘the pricing structure among primary energy sources’ and ‘the current situation on the market for CO2 certificates’ may explain the present lack of investments in conventional electricity generation.6 In addition, currently there is actually no shortage of generation capacity, neither in Germany nor in Europe. Some operators complain of an extremely low electricity price level caused by an ‘overly high share of base-load and mid-merit power plants’ and by ‘overcapacity in the coupled European energy market’.7 With regard to gas-based power plants, the current market structure appears to be problematic: Traditionally high priced gas generated power was used to cover peak-load demand. However, the traditional peak demand times in particular around noon time have been levelled by the high production of photovoltaic renewable generation in particular in the middle of daytime. Some gas-fired power plants were only used for a few hours during the last year and therefore it was not possible to operate them economically.8 The north-south-imbalance in generation capacity9 is closely linked to the nuclear phase-out decided in 2011. Only six out of former eleven nuclear power plants are still running in Southern Germany. According to BMWi10 in the south of Germany, in 3

Energy Concept (n 2) p 4 ff. Deutscher Bundestag, Entwurf eines Gesetzes zur Neuregelung des Rechtsrahmens für die Förderung der Stromerzeugung aus erneuerbaren Energien (Drucksache 17/6071, 6 June 2011) http://dip21.bundestag. de/dip21/btd/17/060/1706071.pdf, accessed 1 February 2015. 5 Erneuerbare-Energien-Gesetz 2012, 25 October 2008 (BGBl. I S. 2074) (EEG 2012), repealed by EEG 2014, 21 July 2014 (BGBl. I S. 1066) (EEG). 6 BMWi, Report of the Power Plant Forum to the Federal Chancellor and the Minister-Presidents of the Länder (28 May 2013) p 6. 7 BMWi, Report of the Power Plant Forum (n 6) p 6. 8 BMWi, Report of the Power Plant Forum (n 6) p 6. 9 BNetzA, Bericht zum Zustand der leitungsgebundenen Energieversorgung im Winter 2012/13 (20 June 2013). 10 BMWi, Report of the Power Plant Forum (n 6) p 5. 4

15.2 Setting the scene

273

particular during peak demand periods (during winter months) the level of generation available at market prices may not always be sufficient to meet demand. This situation is expected to continue through to 2015. The shortage of generation capacity in the south of Germany goes along with abundant renewable energy generation in the north of Germany. Finally, in particular in the run-up to the German federal elections in 2013 it has become quite clear that the cost of the current German RES support scheme is not considered politically acceptable anymore. This scheme costs German electricity consumers more than €20 billion every year. In addition, in December 2013 the EU Commission has opened state aid proceedings against the EEG 2012, the German Renewables Act of 2012 and an important element of the German RES support scheme, the exemption of energy intensive industry from a part of the renewables levy (EEGUmlage).11 The ‘question of a regulatory framework for the future electricity market is inextricably linked to decisions surrounding the fundamental reform of the [ . . . ] EEG.’12 This reform is one of the projects with the highest priority for the new German government, the so-called ‘grand coalition’. It must succeed in order to restore lost confidence in the legislation and to strengthen the willingness to invest in and to avoid EU sanctions. Even though there has been a reform of the EEG in 2014 it is clear that this will not be sufficient but that a more fundamental reform will be required in 2016–2017.

15.2.2 Market characteristics As described earlier, Germany’s current electricity market is an energy-only market with an additional network reserve. In December 2012 the total capacity of connected generating facilities in Germany was 178.3 GW, with 102.6 GW from conventional generating facilities and 75.6 GW from renewable generating facilities.13 The gross power production in Germany was 617 TWh in 2012 (634 TWh in 2013)14 with 542.05 TWh from conventional generating facilities and 135.12 TWh from renewable generating facilities,15 whereas the gross electricity consumption was 606.7 TWh in 2012 (599.8 TWh in 2013).16 Eight nuclear power plants were decommissioned in 2011 (with a total capacity of 8422 MW), the remaining nine nuclear power plants (with a total capacity of roughly 12 GW)17 will be decommissioned between 2016 and 2022.

11 Commission opening decision of 13 December 2013 in Case SA.33995 (2013/C) (ex 2013/NN) Germany—Support for renewable electricity and reduced EEG-surcharge for energy-intensive users its decision to open an in-depth investigation (followed by Commission decision of 25 November 2014) C (2014) 8786 final; now under appeal. 12 BMWi, Report of the Power Plant Forum (n 6) p 18. 13 BNetzA and FCO, Monitoringbericht 2013 (June 2014) p 17. 14 BDEW, Erneuerbare Energien und das EEG: Zahlen, Fakten, Grafiken (2014) (Berlin, 24 February 2014) p 16. 15 BDEW, Erneuerbare Energien und das EEG: Zahlen, Fakten, Grafiken (2013) (Berlin, 31 January 2013) p 15 and own calculation on basis of this data. 16 Hans-Wilhelm Schiffer, ‘Deutscher Energiemarkt 2013’, et 64 (3) (2014) pp 66 and 72. 17 BMWi, Report of the Power Plant Forum (n 6) p 4.

274

Germany

In 2013 there was an 11.8 per cent share of renewable energies in total final energy consumption (12.6 per cent in 2012).18 The share of renewable energies in electricity consumption was 25.4 per cent in 2013 (23.6 per cent in 2012).19 In 2013, once again, the wind energy provided most of the power created by renewables, followed by biomass and photovoltaic.20 With the exception of a few municipal utilities owning renewables capacities all this generation capacity is privately owned. Germany’s electricity grid is divided into transmission grids (extra-high voltage) with four transmission grid operators and distribution grids (high voltage, medium voltage and low voltage) with more than 700 local operators. Traditionally, the electricity was centrally generated by large power plants which had been located relatively close to the centres of consumption and fed the electricity into the transmission grid. The new structures developing for electricity generation from renewable sources have relatively small generation units, many of them solar power panels installed on roofs of private houses which are feeding decentralized electricity into the distribution grids. In addition, a lot of wind power is generated in northern Germany while centres of industrial consumption are in the Ruhrgebiet and in Southern Germany. Therefore, to ensure a reliable energy supply in the future, an adaption and expansion of the grid is urgently needed. In particular, the German transmission network needs to be expanded and reinforced in order to transport electricity from the north to the south of Germany, where the annual electricity demand makes up nearly a third of country’s total electricity demand.21

15.2.3 Regulatory framework In Germany, the main legal acts governing the electricity sector are the Energy Industry Act (Energiewirtschaftsgesetz—EnWG)22 and the Renewable Energies Act (ErneuerbareEnergien-Gesetz—EEG)23 and their related ordinances. The main actor regarding the regulation of the energy market is the Federal Network Agency for Electricity, Gas, Telecommunications, Post and Railway (Bundesnetzagentur—BNetzA, energy regulator). Based on EU directives,24 the German markets for electricity and gas were liberalized by the EnWG in 1998, and were substantially reformed in 2005, 2011, and 2012. Third parties were granted a right to non-discriminatory access to electricity and gas grids, while integrated energy supply companies were forced to unbundle their grid operations 18

BMWi, Erneuerbare Energien in Zahlen im Jahr 2013 (October 2014) p 7. BMWi, Erneuerbare Energien (n 18) p 7. 20 BDEW, Erneuerbare Energien und das EEG (n 14) p 16. 21 Agora Energiewende, Load Management as a Way of Covering Peak Demand in Southern Germany (May 2014) p 4. 22 Energiewirtschaftsgesetz, 7 July 2005 (BGBl. I S. 1970, 3621), last amended in 2014 by BGBl. I S. 1066 (EnWG). 23 EEG (n 5). 24 Directive 96/92/EC of the European Parliament and of the Council of 19 December 1996 concerning common rules for the internal market in electricity [1997] OJ L 27 (1996 Electricity Directive) and Directive 98/30/EC of the European Parliament and of the Council of 22 June 1998 concerning common rules for the internal market in natural gas [1998] OJ L 204/1 (1998 Gas Directive). 19

15.2 Setting the scene

275

from their generation and supply activities and grid access. BNetzA was introduced as a regulator. To ensure the security of energy supply, a strategy was developed which establishes a mechanism to procure a network reserve by compensating participating operators of existing electricity generation or storage facilities. In order to meet the ambitious national target of generating 35 per cent of its electricity from renewable energy sources by 2020, Germany introduced the EEG, which was also revised in 2011 and—after strong and widespread demands for a reform due to the constantly increasing costs of supporting renewables25—in 2014. It provides for a duty to connect and off-take on the part of grid operators, as well as for fixed remuneration tariffs for every kilowatt per hour produced. The priority access to transmission and distribution grids as well as the twenty years of having fixed feed-in tariff have been a very successful system to promote renewables. BNetzA, the German energy regulator, is a separately established senior federal public institution (Bundesoberbehörde) under the responsibility of BMWi. In the area of energy, BNetzA ensures compliance with the EnWG, the EEG and their respective ordinances. It has a task of providing an effective and genuine competition in the supply of electricity and gas, and efficient and reliable operation of energy supply systems in the long term as well as the implementation and enforcement of European energy law. BNetzA is also responsible for implementing efficient approval proceedings to adapt the German extra-high voltage network to the growing use of renewable energy sources.26

15.2.4 Generation adequacy According to an official report of a forum set up by BMWi and the Federal Government’s response of 7 February 2013 to the Commission’s 2012 consultation on generation adequacy (discussed in section 1.4.5 above),27 on account of new facilities that will have gone into operation, a surplus of around 4 GW in the power balance is expected in 2014 and 2015. However, this assumes that the power plant projects shall be finalized (some are still in the planning and construction stage) and no more plants may be closed down. For the time after 2015 the shutdown of the remaining nuclear power plants by 2022 with a total capacity of 12 GW is prescribed by law. Apart from that there is a lot of uncertainty concerning the economic situation of power plants and the construction of new power plants. A decisive element in this development will be the effect of the reform of the EEG and whether there will be changes to the current market design. The problem is perceived as both the lack of security of supply and an inflexible system. The lack of security of supply is closely linked to the decision of the nuclear phase-out as described in section 15.2.1. The currently existing overcapacities cannot 25

BMWi, Report of the Power Plant Forum (n 6) p 18. EnWG (n 22) Art 29, 30 ff, 54 ff, 65 ff; EEG (n 5) Art 61. 27 BMWi, Report of the Power Plant Forum (n 6) p 4 ff. See page 2 of the response, which is available at the Commission’s website at http://ec.europa.eu/energy/en/consultations/consultation-generation-adequacycapacity-mechanisms-and-internal-market-electricity, accessed 1 February 2015. 26

276

Germany

remedy the strained grid situation experienced in the south. ‘In this respect any relief can only be expected with the implementation of the planned north-south electricity routes’.28 However, despite a new planning regime their finalization will take time. As long as there is no solution for the efficient storage of electricity a more flexible electricity generation system and demand side management are needed to compensate the fluctuating output from renewable generation. It is often argued that more energy being generated from renewable sources does not necessarily lead to a lessening in the conventional generating capacity required. Along those lines it is argued that 80–90 per cent of conventional capacity will still be needed in the future,29 but has to be adjusted to the growing output from renewable generation. However, the future demand for conventional generation capacity will depend on a number of developments such as the ability of market mechanisms to combine different sources of renewable energy to provide secured capacity instead of fluctuating power from renewable energy sources, and the development of electricity storage technologies and of interconnection capacities30 between the various European markets. The resulting insecurity adds to the reluctance to currently invest in conventional generating capacity.

15.3 Energy-only market and network reserve 15.3.1 Background As described earlier, the German energy market is currently characterized by an energy-only market with additional strategic reserves to protect against regional and seasonal shortages, called in the terminology of the German Energiewende a ‘network reserve’.31 Germany introduced the facility of the network reserve in 2013 after it experienced severe regional capacity shortages, in particular in Southern Germany. The aim was to ‘create the framework conditions that are needed in the short term to help prevent any risk to supply security without already defining the blueprint for a new “electricity market design”’.32 The facility mainly consisted of an amendment of the EnWG and an ordinance together with accompanying measures by the German regulator BNetzA. Also at the end of 2012 the obligation to notify the responsible transmission system operator (TSO) of a temporary or final decommissioning of plants 28

BMWi, Report of the Power Plant Forum (n 6) p 5. BDEW, BDEW-Projektgruppe Kapazitätsmechanismen, Zwischenbericht an den BDEW-Vorstand (25 October 2011). 30 The German grid is interconnected with all bordering countries, especially with France, the Netherlands and Switzerland. See http://www.bmwi.de/DE/Themen/Energie/Strommarkt-der-Zukunft/zahlenfakten.html, accessed 1 February 2015, for a map provided by ENTSO-E, Deutschland im europäischen Netzverbund; BDEW, Stromaustausch Deutschlands mit dem Ausland (14 January 2014) p 2; section 15. 31 The German government introduced two different kinds of reserve capacities that are most of the time summarized with the term ‘network reserve’. The network reserve that has its basis in the Reservekraftwerksverordnung (ResKV) of 27 July 2013, BGBl. I S. 1947 Art 2–9 (section 15.3.2.2) and the here so-called ‘strategic network reserve’ introduced by the EnWG (n 22) Art 13a–13c (sections 15.3.2.3 and 15.3.2.4). Whereas the nature of the first one is more short term, the latter strategic network reserve is for longer periods and therefore kind of a strategic reserve. 32 BMWi, Report of the Power Plant Forum (n 6) p 5. 29

15.3 Energy-only market and network reserve

277

at least twelve months in advance was introduced in order to ensure the security of supply in the near future. The duration of these measures as well as of the rules on the network reserve are limited to 31 December 2017. German energy law currently also allows for the introduction of capacity mechanisms. However, this is permitted only under the restrictive circumstances of Article 53 EnWG. The introduction of alternative/supplementary legal instruments is currently being discussed.

15.3.2 Key features 15.3.2.1 General measures by TSOs The TSO may solve temporary difficulties in the market with the help of short-term market-management. It is allowed to resolve a danger or disturbance of security of the supply of electricity with the help of market-related measures (marktbezogene Maßnahmen), such as, in particular, the use of balancing energy, contractually agreed upon loads that can be turned on and off, information regarding congestion, management of congestion by way of redispatch, and the mobilization of additional reserves. The TSO may request the operators of power generation units and storage facilities with a power rating above 10 MW to adjust their use of active and reactive power (Wirkleistungseinspeisung and Blindleistungseinspeisung) for a reasonable compensation (redispatch). There is a controversial discussion about the relationship between the rules of redispatch and the rules concerning the reimbursement for temporarily decommissioned facilities.33 The discussion is between those who consider redispatch to be a sufficient means of response to transmission bottlenecks and those who consider strategic reserves in regions with a potential generation capacity shortage to be required.

15.3.2.2 Network reserve With the introduction of the so-called Ordinance on Reserve Power Plants (Reservekraftwerksverordnung—ResKV)34 the practice between TSOs and generators of using a few generation units as a network reserve (from winter 2011–2012 on) on the basis of contracts was codified. Rules on the network reserve were adopted on 27 June 2013 and entered into force on 6 July 2013. The ordinance has been introduced for a limited time and shall remain in force until 31 December 2017. The ResKV regulates the procedure in relation to the creation of a network reserve from existing or, as an exception, also from new facilities in order to ensure the safety and reliability of the electricity supply system. The creation of a network reserve is based on agreements between the TSO responsible for the respective region and the 33 Deutscher Bundestag, Antwort der Bundesregierung auf die Kleine Anfrage der Abgeordneten Oliver Krischer, Cornelia Behm, Bettina Herzlitzius, weiterer Abgeordneter und der Fraktion BÜNDNIS 90/DIE GRÜNEN (Drucksache 17/14733, 11 September 2013) p 3. 34 ResKV (n 31). The German federal government introduced the ResKV pursuant to EnWG (n 22) Art 13b(1)–(2).

278

Germany

operator of the respective power plant. The agreement has to be negotiated in consultation with the regulator BNetzA. The creation of a network reserve takes priority over a shutdown of generating facilities planed by the operator. The procedure to create a network reserve is set out in Articles 3 to 5 of the ResKV. According to these provisions, BNetzA annually reviews the need for generation capacity for the network reserve by 1 May every year. This review is based on the so-called system analysis carried out by all TSOs together in terms of the available generation capacity, its development with regard to the next winter season, the estimated needs for the next five years, and the possible need for network reserve. Once BNetzA has confirmed the need for generation capacity, the relevant TSO publishes the exact requirements on generation facilities including place and technical parameters by 1 May every year. Plant operators may express their interest in concluding a contract by 15 May every year. Contracts have to be concluded by 15 July every year. BNetzA has to be consulted in this process. In case of two identical offers (with the same technical suitability), the TSO has to choose the cheapest offer. The agreements may run up to twenty-four months (or longer in particular cases). The agreement has as a prerequisite that the generating facility has to be system-relevant. A plant is considered to be system-relevant if the final shutdown affects the security of supply/the reliability of the electricity supply system in a significant way and if this danger/disturbance of supply cannot be contained by any other means (see below, section 15.3.2.4). Furthermore the agreement requires the commitment to waive the right to use the plant in the energy market after expiry of the contract, the expiry of the notification period,35 as well as relevant approvals. The TSO may also enter into a contract with operators of power plants located in the EU or Switzerland if the generating facility is able to contribute to the German security of supply, provided that the relevant authorities of the other state don’t raise any objections. Furthermore this requires that the cross-border supply is secured and that the generating facility is at least as technically suitable and not more expensive than generating facilities in Germany. The network reserve is excluded from the energy market and may only be used in coordination with the TSO. The amount of reimbursement for the use of the network reserve is determined by the price for feed-ins, by the costs for restoration of operational readiness, by the power price including the contingency costs, and by the recognition of overheads (5 per cent of other expenses). There is neither reimbursement for costs that would also have been incurred had the facility been shut down nor compensation for opportunity costs. The TSO may add all costs related to such an agreement to its network fees and charge them to the grid customer. In exceptional circumstances new generation facilities may also be acquired as network reserve. In this case BNetzA has to confirm the relevant need. Construction and operation of such a generating facility have to follow a strict tendering procedure

35

Pursuant to EnWG (n 22) Art 13a(1), operators of power generation units and storage facilities with a power rating above a certain amount have to inform the responsible TSO and BNetzA about a temporary or final decommissioning of their plant at least twelve months in advance (see section 15.3.2.3).

15.3 Energy-only market and network reserve

279

(in exceptional cases the TSO itself may, in accordance with BNetzA, construct and operate such a new facility).36

15.3.2.3 Notification requirement and the twelve-month moratorium on the closure of plants In order to ensure the system stability, operators of power generation units and storage facilities with a power rating above 10 MW have to inform the responsible TSO and BNetzA about a temporary or final decommissioning of their plant/parts of their plants at least twelve months in advance.37 If the continued operation is still legally and technically possible it is forbidden to shut down a plant during this twelve-month period. During this period the plant operator is obliged to keep the plant ready to start generation. A facility may be shut down earlier than twelve months after the notification, if the TSO has confirmed that the shutdown has no influence on the system security or reliability of the energy system.38 During the twelve-month period the TSO is entitled to requisition the plant’s operation if this is necessary for the safe and reliable operation of the grid.39 If the TSO orders the plant operator to run the plant, the plant operator has a right to adequate compensation.40 The compensation includes reimbursement of production expenses (Erzeugungsauslagen) as well as expenses for operational availability (Betriebsbereitschaftsauslagen). There is no reimbursement of costs that would have been incurred anyway as well as of opportunity costs. If the plant operator claims compensation for its expenses for operational availability, it may only operate at the TSO’s disposition for five years. The plant operator has to refund this compensation, ie the residual value of investment advantages that it has drawn from the compensation (investive Vorteile), if he returns to the energy market after five years. The TSO is allowed to add its costs to the grid fees.

15.3.2.4 No final shut down of system relevant generation and storage units Also, after this twelve-month period a power plant or a power storage facility with a power rating of more than 50 MW may not be shut down if the responsible TSO does consider it system relevant and BNetzA approves such an assessment. In case of notification of a final shutdown of such a large plant, the responsible TSO subsequently has to assess the ‘system relevance’ of the plant. As mentioned, a plant is considered to be system relevant, if the final shutdown affects the security of supply/the reliability of the electricity supply system in a significant way and if this danger/disturbance of supply cannot be resolved by any other means. A plant may be designated as system relevant for a maximum of twenty-four months. After this period the responsible TSO has to assess again whether the plant continues to be system relevant. If the facility is deemed system relevant by the TSO with BNetzA’s approval and a continued operation is technically and legally possible, final shutdowns of plants with a 36 39

ResKV (n 31) Art 8. EnWG (n 22) Art 13.

37

EnWG (n 22) Art 13a. EnWG (n 22) Art 13(1b).

40

38

ResKV (n 31) Art 10(5).

280

Germany

nominal capacity of 50 MW or more are prohibited also after the twelve-month period. In such a case where permanent closure is prohibited, the plant operator must maintain the plant in a condition that allows a request for further provision or restoration of operational readiness, provided it is technically and legally possible. The plant operator is entitled to a reasonable compensation for the necessary conservation measures (socalled maintenance expenses) from the responsible TSO. In this case until its final decommissioning the plant or the storage facility may only be operated according to the system security measures ordered by the TSO. By the end of October 2014 plant operators had informed BNetzA about the intended final decommissioning of thirty-two generation facilities:41 eleven facilities have already been considered systemically relevant by the TSOs, four have already been shut down.

15.3.2.5 Fuel prerogatives of large gas-fired power plants After the shutdown of nuclear power plants in March 2011 during the winter of 2011–2012 southern Germany could only be adequately supplied with the use of the strategic network reserve.42 In addition, a gas supply crises endangered the gas supply of some system relevant gas power plants in February 2012.43 In order to prevent such situations from occurring in the future the TSOs responsible were entitled to designate gas-fired power plants with a power rating of more than 50 MW as system relevant.44 This designation requires BNetzA’s approval and can be for a maximum period of twenty-four months. As a consequence the TSO can order the relevant gas transmission system operator not to restrict the supply of gas to strategically relevant gas-fired power plants during periods of gas shortage (Art 16(2a) EnWG). At the same time the operator of the gasfired power plant is obliged to change the fuel burnt by the power plant, for example, to switch to oil as a fuel, if there is the reasonable technical, legal, and economic possibility to do so. In this case, the power plant operator will be reimbursed for its additional costs.

15.3.2.6 Article 53 EnWG In Article 53, the German Energy Industry Act also allows for the introduction of other capacity mechanisms.45 However, this is permitted only under restrictive circumstances. The Federal Government can adopt a new ordinance introducing a tendering procedure or an equivalent non-discriminatory procedure for the acquisition of new capacities or for energy efficiency measures or demand side instruments. This requires that the security of supply cannot be ensured through existing generation capacities or 41 BNetzA, List of notifications concerning planned shutdowns (Kraftwerksstilllegungsanzeigenliste) (20 October 2014). 42 Christian König, ‘Art 13c EnWG’ in Franz-Jürgen Säcker (ed), Berliner Kommentar zum Energierecht (vol 1, part 1, Deutscher Fachverlag GmbH, 2014). 43 44 45 König (n 42). EnWG (n 22) Art 13c. EnWG (n 22) Art 53.

15.3 Energy-only market and network reserve

281

energy efficiency and demand management measures. Such an ordinance can only be adopted with the consent of the Federal Council (Bundesrat). The Federal Government has, to date, made no use of this option. There is an ongoing debate among jurists about whether this provision allows for the introduction of a capacity mechanism just by way of ordinance of the Federal Government. Some proposals for a capacity mechanism may be covered by this provision.

15.3.3 Enforcement The acquisition of balancing and reserve power is generally based on contracts between TSOs and plant operators. However, BNetzA plays a key role enforcing compliance with the legal requirements as it is involved in many steps in the described procedures. First, BNetzA approves plants deemed system relevant by the TSO. Second, the agreements between TSOs and plant operators may only be made in coordination with BNetzA. Third, BNetzA also reviews, together with the TSOs, the need for a network reserve. In that respect, BNetzA determined that in Winter 2014–2015 there was a need of reserve capacity of 3091 MW of which 3027 MW are contractually secured, whereas in Winter 2015–2016 a reserve capacity of 6000 MW will be required of which 3225 MW are contractually secured.46 Thus BNetzA is at a central position to influence and later to control the development in order to determine whether a capacity mechanism is required in Germany or not. More and more BNetzA is also developing into the German government’s institution which has the most extensive personnel resources and expertise for developing the Energy Transition so that the government increasingly relies on BNetzA for its assessment of problems and their resolution.

15.3.4 Alternative measures Despite these legal instruments of the two kinds of strategic reserves presently existing in Germany, the introduction of alternative/supplementary legal instruments is being discussed. It is argued that the energy-only market does not send sufficient long-term investment signals to ensure the security of supply through construction of conventional generation capacities in order to cope with the fluctuating nature of renewable energies.47 In addition, the urgency of the problem is increased by the fact that by 2022 the nuclear generating capacity amounting to roughly 12 per cent of the current conventional German generation will be shut down. Some suggestions propose a strategic reserve while others prefer introducing a capacity mechanism.48 Others, a minority, count on the market forces in the energy-only market.

46 BNetzA, Feststellung des Reservekraftwerksbedarfs für den Winter 2013/14 sowie die Jahre 2015/2016 und 2017/2018 und zugleich Bericht über die Ergebnisse der Prüfung der Systemanalysen (2 May 2014), pp 11, 59, 75. 47 BMWi, Report of the Power Plant Forum (n 6) p 7. 48 BMWi, Report of the Power Plant Forum (n 6) p 7.

282

Germany

15.3.4.1 Federal Government The Federal Government recognizes the need for more market based mechanisms and competition in the electricity market to ensure cost-effective, reliable, and low-carbon electricity supply. The question of a regulatory framework for the future electricity market is, according to the Government, ‘inextricably linked to decisions surrounding the fundamental reform of the EEG’ and it is also linked to the ‘issue of CO2’, ‘as carbon prices also affect the number of hours gas power stations operate’.49 Also after the 2014 reform of the EEG, this statement remains correct as the next reform is already announced for 2016. Pursuant to the coalition agreement dated 27 November 2013 the government’s assessment is that Germany currently has a sufficient number of power plants. However, in the medium term a capacity mechanism shall be developed, considering the aspect of cost efficiency, consistent with European rules and ensuring a competition oriented solution open for different technologies.50 In the long term a decision between an ‘optimized electricity market (electricity market 2.0)’ and the introduction of a ‘further market alongside the electricity market for the maintaining of reserve capacity (capacity mechanism)’ has to be made. In May 2013 BMWi proposed to defer any decisions about ‘far-reaching measures’ and to first thoroughly review, eg by monitoring, market simulation, impact-assessment of capacity mechanisms, and stocktaking, the issues that still remain open.51 Subsequently the Ministry commissioned two studies—a study on performance of the energy-only market52 and an Impact Assessment on capacity mechanism53—which both seem to favour the development of an optimized electricity market. According to a 10-point-energy-agenda54 and on the basis of several studies and discussions with relevant parties the Ministry released a Green Paper (Grünbuch)55 discussing different future energy market designs. Public discussions shall be followed by a White Paper (Weißbuch) that proposes concrete measures at the end of May 2015. These measures will also be publicly discussed until September 2015. Necessary legislative actions are announced to be introduced still in 2015. This project goes along with a dialogue with the neighbouring states as well as the Commission. The network reserve as a temporary, safeguard solution will be further developed and executed until a final decision has been made.

49

BMWi, Report of the Power Plant Forum (n 6) p 18f. Coalition Agreement between CDU, CSU and SPD (27 November 2013) p 57. 51 BMWi, Report of the Power Plant Forum (n 6) p 18 ff. 52 Frontier Economics et al, Strommarkt in Deutschland—Gewährleistet das derzeitige Marktdesign Versorgungssicherheit? (July 2014). 53 Frontier Economics et al, Folgenabschätzung Kapazitätsmechanismen (Impact Assessment) (July 2014). 54 BMWi, Zentrale Vorhaben Energiewende für die 18. Legislaturperiode (10-Punkte-Energie-Agenda des BMWi) (26 June 2014) p 6. 55 BMWi, Ein Strommarkt für die Energiewende (n 1). 50

15.3 Energy-only market and network reserve

283

A study on behalf of the Ministry for the Environment, Nature Conservation and Nuclear Safety dated from 12 December 2013 proposes to introduce a strategic reserve in case an advancement of the energy-only market is necessary.56

15.3.4.2 German Association of Energy and Water Industries (BDEW) In a statement published at the end of September 2013, immediately after the German federal elections, the German Association of Energy and Water Industry (BDEW) together with the Association of Municipal Enterprises (Verband kommunaler Unternehmen—VKU) demanded a fundamental reform of the EEG. They proposed the introduction of (a) a decentralized capacity mechanism and (b) strategic reserve.57 In BDEW’s proposal capacity certificates are issued which guarantee availability of a certain capacity under all circumstances. Generators offer those certificates according to the capacity available to them and balance group managers buy them for their consumers according to their capacity needs. The price for those certificates will be centrally determined by way of a market mechanism. According to BDEW, such a model would ensure the economically viable operation of existing conventional facilities as well as investments in new generating facilities. Additionally, a strategic reserve, similarly to the existing network reserve, shall ensure the energy supply. Finally, BDEW demands for the integration of the Energy Transition into the European electricity and gas market. This proposal, which is supported by two large and powerful industry associations, has a broad basis in the German energy industry and in the political parties. (a) Decentralized capacity mechanism BDEW’s proposal for the introduction of the decentralized capacity mechanism58 is in addition to the already existing energy-only market. The details are as follows. The proposal provides for the introduction of special certificates (so-called Versorgungssicherheitsnachweise—VSN) that represent a certain amount of guaranteed/secured energy, ie energy for which even in case of an energy shortage the supply is ensured by physical generation capacities in the form of power plants and storage facilities. Balancing group managers (Bilanzkreisverantwortliche) have to secure their required amount of energy in case of energy shortage with the acquisition of these certificates in advance of this shortage from providers of secured energy. Domestic power plants and storages as well as foreign operators may be suppliers of secured energy, provided that they are able to ensure the energy supply to the extent that they have sold the VSN. Further, to ensure the sufficient purchase of VSN, the balancing group manager is subject to penalty if it doesn’t have enough capacity available to satisfy its customers’ needs in case of an energy shortage. It is essential for the VSN-model that in cases of 56 Energy Brainpool, Vergleichende Untersuchung aktueller Vorschläge für das Strommarktdesign mit Kapazitätsmechanismen—Kurzstudie im Auftrag des BMU (12 December 2013). 57 BDEW, Der Weg zu neuen marktlichen Strukturen für das Gelingen der Energiewende (18 September 2013). 58 BDEW, Ausgestaltung eines dezentralen Leistungsmarkts (18 September 2013).

284

Germany

power shortages the volume of energy used is limited to the VSN-ensured energy. It is also essential that the sellers of VSN are available at any time to feed-in the amount of power corresponding to the amount of VSN sold. That implies that in particular the following issues need to be regulated. First, the amount of the penalty, if, in case of power shortage, either more power is used than is secured by VSN or less power is provided than VSN were sold. In addition, the market participants’ compliance with their duties has to be monitored. And finally one needs a definition of power shortage. BDEW proposes to take the level of day-ahead prices as an indicator of power shortage. Other possible indicators which have been proposed, like the maximum load of the customers which are part of a balancing group, actual compared to booked, will only be known ex post. Furthermore, in order to avoid speculations on the nonoccurrence of the ‘shortage-price’ sellers of VSN and balancing group managers have to prove that at any time they are able to fulfil their obligations.59 Otherwise security of supply is at risk in case of a shortage of supply. In the VSN-model the demand for guaranteed energy is not regulated through a ‘planned economy’ approach but is demand-based, ie the need for security of supply is determined by the demand side and therefore by the customer, as final consumer. BDEW also proposes that the customer may additionally commit himself to adapt its demand according to circumstances (demand flexibility). (b) Strategic reserve In addition, BDEW proposes the introduction of a strategic reserve as a kind of ‘safety net’ that shall support the transformation of the energy system. A study60 conducted by BDEW, BEE (German Association for Renewable Energies), and several scientists published in 2013 proposes this solution. In the system proposed by BDEW a strategic reserve would replace the network reserve over the next few years.61 The key components of BDEW’s proposals for an additional strategic reserve are the following. The use of the strategic reserve would apply on the day-ahead market of the electricity exchange (EPEX Spot), if otherwise the demand cannot be met at the maximum permitted price in the second round of auction which is currently 3000 €/MWh. Furthermore, the use of the strategic reserve should be restricted to redispatch, for example to ease the grid situation in Southern Germany. The acquisition and use of plants as a system reserve by the TSO would be under the control of the Federal Ministry of Economic Affairs and Energy and BNetzA. The volume of the strategic reserve would be determined by BNetzA in cooperation with the TSOs, eg 5 per cent of the annual peak-load. The strategic reserve would be provided by way of tender and would be open to existing and, with some restrictions, to new power plants. New power plants should only be able to participate if they are located in a region that needs additional support (eg Southern Germany). Remuneration would be received in the 59 Either through provision of less energy than guaranteed through VSN (seller) or through buying less VSN than required in case of a shortage (buyer). 60 BMU, BDEW, BEE et al, Märkte stärken, Versorgung sichern, Konzept für die Umsetzung einer strategischen Reserve in Deutschland, Ergebnisbericht des Fachdialogs ‘Strategische Reserve’ (May 2013). 61 BMU, BDEW, BEE et al, Märkte stärken (n 60); BMWi, Report of the Power Plant Forum (n 6) p 9 ff.

15.4 European dimension

285

form of a service price based on the bid and an electricity charge per KWh based on the actual operating costs. A redispatch would be ordered by the TSO responsible for the region and would be remunerated on the basis of the cost-based rules set by BNetzA. The costs of the strategic reserve would be recovered by the TSO via the grid charges that it charges to the customers in its region.

15.3.4.3 Position of the European Energy Exchange (EEX) According to a position paper published by the EEX in 2013,62 the EEX is not convinced that considering the European context a capacity mechanism is required to ensure the security of supply in Germany. The proposed mechanisms are considered to be too complex, too expensive, and to have too many problems. The EEX proposes that instead of introducing a capacity mechanism, other measures should be taken aiming at further development of the European internal market, promotion of energy efficiency, demand side management, and an integration of renewables into the energy market. If a capacity mechanism should prove unavoidable, necessary, the EEX proposes the introduction of a capacity market scheme if there is still a need for new generation facilities or insufficient willingness to invest. The EEX strongly favours a strict decoupling between any capacity mechanism and the energy market. A capacity mechanism should only be introduced to ensure the security of supply instead of pursuing multiple targets (eg reduction of CO2-emissions, integration of renewables etc). Furthermore, the EEX proposes a European solution, at least a solution aligned on a regional European level (eg within the CWE-region). In addition it suggests that it should be ensured that in total the Member States do not build up too much capacity by establishing separate national capacity mechanisms. The EEX also proposes an equal treatment of existing and new generating facilities and clarifies that regional measures to address network congestion problems are only a temporary solution. If a capacity mechanism is considered necessary at all in order to ensure the security of supply, the EEX prefers a decentralized capacity mechanism to a centralized model.63 It also rejects the introduction of a long-term strategic reserve. This would only be temporarily useful.

15.4 European dimension The current energy supply system in Germany, consisting of an energy-only market with an additional network reserve, has cross-border implications: First, already at present there are almost thirty foreign generating facilities larger than 10 MW that ensure the German energy supply every day.64 Second, even before the ResKV65 was 62 EEX, Positionspapier der European Energy Exchange AG, Notwendigkeit und Design von Kapazitätsmechanismen (5 August 2013). 63 For the decentralized model, see section 15.3.4.2 (a). In a centralized model the required capacity is sold through a centralized auction (see EEX, Positionspapier (n 62) p 5). 64 BNetzA, Register of generating facilities (16 July 2014). 65 See section 15.3.2.2 above.

286

Germany

introduced, foreign generating facilities from Austria, Italy, and Switzerland concluded contracts with German TSOs on the provision of reserve capacities.66 Third, the ResKV explicitly provides for the possibility to contract necessary capacities from generating facilities in the Internal Energy Market and Switzerland.67 According to this the calls for expressions of interest of the German TSOs are not restricted to national generating facilities. Generators with facilities located in Austria, Italy, Switzerland, France, the Czech Republic, and other European countries may also express their interest for the inclusion of electricity generation or storage facilities in the network reserve.68 Concerning the future market design of the German energy supply system, the approach of the German Federal Government seems to be in line with the November 2013 Communication.69 According to the Commission, the introduction of a capacity mechanism requires ‘an objective, facts-based, and comprehensive assessment of the generation adequacy situation’ prior to public intervention. In case alternative measures, such as the ‘promotion and enabling of demand response, including by an accelerated roll out of smart metering and [the] expansion of interconnection capacity’, do not solve the problem of generation adequacy, the Commission prefers ‘a strategic reserve’ or ‘a credibly one-off tendering procedure’ to ‘a market-wide capacity mechanism’.70 Finally, as discussed earlier71 the German Federal Government both from the ministerial side as well as the new government under the ‘grand coalition’ supports a European solution as well as regional solutions. It proposes a ‘coordination in suitable regional committees (eg Pentalateral Forum) to work towards the further enhancement of the electricity market design including capacity mechanism, where applicable, to be able to establish a country-specific level of reliably available capacity which, in addition to load and generation, also takes trading into account.’72 According to the Federal Government,73 the cross-border electricity trade is considered important to ensure the security of supply. Therefore the Europe-wide grid expansion as well as an increasing coupling of the markets should be intensified. Steps taken at European level should ensure the security of supply in every Member State. This shall include the further integration of markets, full market liberalization, removal of regulated prices, and developing the flexibility of the demand side (load management). However, the German Government recognizes that given that the debate about the need of a capacity mechanism is still under way, it is too early to set out a blueprint for an EU-wide capacity mechanism. 66 Staatsministerium Baden-Württemberg, Energiewende auf den Punkt gebracht, available at http:// www.baden-wuerttemberg.de/de/bw-gestalten/nachhaltiges-baden-wuerttemberg/energie/energiewendeauf-den-punkt/, accessed 1 February 2015. 67 ResKV (n 31) Art 5(3). 68 Eg TRANSNET BW, Call for expressions of interest, available at http://www.transnetbw.de/de/ strommarkt/systemdienstleistungen/netzreserve, accessed 1 February 2015. 69 Communication from the Commission, Delivering the internal electricity market and making the most of public intervention, C(2013) 7243 final (November 2013 Communication). Commission staff working document, Generation Adequacy in the internal electricity market—guidance on public interventions, 5 November 2013 (Generation Adequacy SWD). 70 71 November 2013 Communication (n 69) pp 13–14. See section 15.3.4.1. 72 BMWi, Report of the Power Plant Forum (n 6) p 19. 73 Response of the Federal Government (n 27) pp 2 ff and 9.

15.5 Conclusion

287

15.5 Conclusion Germany now has to decide how to achieve a long term secure, affordable, and balanced energy supply in the context of the Energiewende. There is consensus that the current measures of a network reserve are not a final solution. On the other hand a balanced generation structure very much depends on the further development of the support system for renewable energy generation and on the European rules for capacity mechanisms. In both respects the current development of state aid rules plays an important role, but the European solution is not yet clear. Nonetheless, with the green paper of November 2014, the German government has started an open discussion on the future market design. At the moment the ‘electricity market 2.0’ seems to be the solution favoured by BMWi, as the introduction of capacity mechanisms would require a higher grade of state intervention in the market and would have to be notified as state aid to the Commission. At the same time, the government wants to talk to the neighbouring countries concerning further cooperation. Following a white paper planned for summer 2015, the government envisages taking legislative measures by the end of 2015.

16 Greece Antonis Metaxas1

16.1 Introduction This chapter introduces the principal characteristics governing the capacity mechanism implemented in Greece aiming to encourage electricity generation and security of supply in the country. The analysis is divided in three sections. Section 16.2 sets out the main characteristics and the legal framework governing the Greek energy market. In section 16.3 the provisions establishing the capacity mechanism in Greece are analysed, with special reference to the proposed amendments by the Greek energy regulator to be implemented in light of the new EU state aid regulatory framework in the energy sector. Lastly, section 16.4 elaborates on the compatibility of the Greek capacity mechanism with EU law.

16.2 Setting the scene 16.2.1 Market characteristics The energy market in Greece was liberalized in 1999 through the adoption of Law 2773/1999,2 which transposed the 1996 Electricity Directive and the 1998 Gas Directive into national law.3 Today, the market continues to be dominated by the incumbent energy company, the Public Power Corporation SA (PPC) which remains vertically integrated, state-controlled and is active either directly or indirectly in all parts of the energy market (production, transmission, distribution, and supply) through its wholly owned subsidiaries: ADMIE (the Hellenic Transmission System Operator, TSO) and DEDDIE (the Hellenic Distribution System Operator, DSO). PPC’s market share, along with its exclusive access to lignite and hydroelectric resources (Greece’s most competitive energy mix for electricity production) makes it difficult for new players to enter the generation market.4 1 The text takes into account legislative developments until 15 October 2014. The author thanks Danai Fati and Areti Kostaraki, both Members of the Hellenic Energy Regulation Institute, for their substantial contribution to this chapter. 2 Law 2773/1999, Government’s Gazette, Issue ` 286 (GG A 286), 22 December 1999. 3 Directive 96/92/EC of the European Parliament and of the Council of 19 December 1996 concerning common rules for the internal market in electricity [1997] OJ L 27 (1996 Electricity Directive) and Council Directive (EC) 98/30 concerning common rules for the internal market in natural gas [1998] OJ L 204/1 (1998 Gas Directive). 4 PPC’s exclusive rights to lignite were recently found to breach Art 106 TFEU in conjunction with Art 102 TFEU by the CJEU on the basis that they create a situation in which a public undertaking or an undertaking on which the state has conferred special or exclusive rights is led to abuse its dominant position: Case C-553/2012 Commission v DEI, judgment of 17 July 2014, nyr. See, to that effect, Case C-462/ 99 Connect Austria [2003] ECR I-05197, para 80 and case law cited therein.

16.2 Setting the scene

289

Due to the special physical geography of Greece and its numerous islands, its electricity grid is divided into the system of the mainland and the network of the islands. The mainland’s system, called the ‘interconnected system’ (which also includes the islands of Western Greece, Andros, Corfu, Lefkada, Kefolonia, and Zakynthos which are interconnected with the mainland through submarine cables) is administered by the TSO whereas the system of the islands, called the ‘non-interconnected network’, is administered by the DSO.5 The relevant electricity market for Greece is, to a significant extent, the national market, as a regional market has not yet emerged.6 The total interconnection capacity in 2011 was 2000 MW and covers all its neighbouring countries, Italy, Bulgaria, Albania, Former Yugoslav Republic of Macedonia (FYROM), and Turkey.7 As a result of market liberalization, a few independent companies are now active in electricity production, which are mainly involved in electricity supply.8 Those active in production have constructed and operate gas-fired power plants, but current high gas prices and high taxation render these units uncompetitive compared with lignite and hydro units.9 All these structural and regulatory constraints are accompanied by ‘the rapid, unplanned and unbalanced RES penetration’10 into the country’s energy mix. Renewables, despite having a number of considerable extremely beneficial effects,11 destabilize the grid due to their sometimes ‘fluctuating, intermittent operation.’12 This situation requires that conventional power plants are available and ready to respond to demand in case of RES production interruption. There has therefore been a need to develop a capacity mechanism which would allow conventional plants to recover their costs. In 2005 Greece adopted a scheme based on capacity obligations (Capacity Assurance Mechanism, CAM), which has almost never been implemented in practice. Instead, in 2006, Greece implemented a temporary capacity mechanism, based on capacity payments, which is practically still in force (Transitional Capacity Assurance Mechanism, transitional CAM). Both mechanisms are discussed in sections 16.3.1 and 16.3.2 below. The Greek energy market is currently undergoing major restructuring, especially regarding the design of its wholesale market and the capacity mechanism. The reforms

5

ADMIE (TSO), see transmission system description, available at http://www.admie.gr/en/ transmission-system/system-data-description/transmission-system-description/, accessed 1 February 2015. 6 RAE, National Report to the European Commission (October 2012) pp 14–18. 7 A detailed map of Greece’s interconnections is available at http://www.admie.gr/en/transmissionsystem/system-data-description/map/, accessed 1 February 2015. 8 The list of registered and licensed companies can be found in LAGIE’s (market operator) website. 9 RAE, Decisions 338 and 339/2013, GG B 1795, 25 July 2013, preamble. 10 For an analysis of the deficits of the Greek policy on the development of RES, see Antonis Metaxas and Michael Tsinisizelis, ‘The development of renewable energy governance in Greece. Examples of a failed (?) policy’, in Michalaina and Hills (eds), Renewable Energy Governance, Lecture Notes in Energy (Springer, 2013), Vol 57, pp 155–68; Antonis Metaxas and Phedon Nicolaides, ‘Asymmetric tax measures and EU state aid law. The special solidarity levy on Greek producers of electricity from renewable energy sources’ (2014) European State Aid Law Quarterly (EStAL) 1, 51–60. 11 Such as the fact that they constitute an environmentally friendly energy source, contributing to the national and EU target of limiting greenhouse gas emissions. For the increased penetration of RES, especially as regards Greece, speak also very important geopolitical reasons and the necessity to support policies that facilitate the energy autonomy of the country by reducing the dependency on imported fossil fuels. 12 RAE, Decision 338/2013 (n 9), preamble.

290

Greece

are necessary to address all the asymmetries and distortions of the rather atypical Greek electricity market as well as to align national regulation with EU law. Since the beginning of the liberalization process, the Greek wholesale market has been organized according to a mandatory day-ahead pool model.13 This model is considered particularly complex and difficult to apply, even in mature, truly liberalized energy markets.14 Despite efforts to introduce transparency into the price-setting process certain units fail to recover their variable costs.15 This has distorted the market and created asymmetries between different producers. This prompted the Greek government to implement a capacity mechanism in order to compensate producers for parts of their losses and incentivize investment in private generating units by adopting a Capacity Assurance Mechanism. As the provisions on the transitional CAM expire in principle on September 201416 and in the framework of the reorganization of the Greek market, the Greek Regulatory Authority of Energy (RAE, energy regulator) is already preparing amendments to the CAM which aim to create incentives to retire inefficient power plants and/or incentivize investments herein.17

16.2.2 Regulatory framework The Greek energy market is currently governed by Law 4001/2011 (amendment to Law 2773/1999)18 which sets the general normative framework for the electricity market, transposing the Third Energy Package19 into national law. This Law entrusts the energy regulator with substantial market supervisory and investigatory powers, as well as commissions it to develop the Grid and Market Operation Code (the Grid Code). The Grid Code currently in force was adopted in 2005 and includes detailed provisions governing the capacity mechanism, both the permanent and the transitional one.20 According to Article 95(4) of Law 4001/2011, the TSO carries out a specialized study on the system’s generation adequacy (Capacity Adequacy Study). In particular, it analyses the system’s available capacity, demand data, and cross-border trade, taking 13 Pursuant to provisions of Law 2773/1999 (now Law 4001/2011 on the Operation of Energy Markets of Electricity and Natural Gas, for Research, Production and transmission networks of hydrocarbons and other regulations, GG ` 179, 22 August 2011). 14 Nikos Vassilakos, ‘National electricity market and Target Model: There’s no turning back’, Greek Energy Magazine (2014), pp 14–21. 15 Vassilakos (n 14) p 14. 16 An extension of the application of the transitional CAM until December is already contemplated by RAE, according to the press (see energypress, at http://www.energypress.gr/news/Eghwria-kai-eyrwpaikhkathysterhsh-gia-ta-nea-ADI, accessed 1 February 2015). 17 RAE’s proposals from November 2012 reflecting the need to introduce critical changes to the Greek energy market can be found at http://www.rae.gr/site/categories_new/about_rae/elecmarktreorg.csp, accessed 1 February 2015 (no English version available). 18 Law 4001/2011 (n 13). 19 Regarding electricity: Directive 2009/72/EC of the European Parliament and of the Council of 13 July 2009 concerning common rules for the internal market in electricity and repealing Directive 2003/54/EC [2009] OJ L 211/55 (2009 Electricity Directive) and Regulation (EC) 714/2009 of the European Parliament and of the Council of 13 July 2009 on conditions for access to the network for cross-border exchanges in electricity and repealing Regulation (EC) 1228/2003 [2009] OJ L 211/15 (2009 Cross-border Regulation). 20 Grid Code, adopted by the Decision No ˜5-H¸/B/ØŒ./8311/09-05-2005 of the Minister for Development, GG B 655, 9 May 2005.

16.2 Setting the scene

291

into account the system’s ten-year Development Plan and the long-term energy design of Greece. In view of the expected implementation of the EU Target Model,21 the energy regulator had put forward a set of proposals reflecting the need to introduce crucial changes to the structure and operation of the national energy market in order to stimulate new entry and competition and to eliminate distortive effects that arise from the dominant position of PPC.22 Among others, a number of transitional regulatory measures were adopted23 in order to amend the capacity mechanism which was adopted at the outset of the liberalization process.24 These reforms, which have been adopted since 2013 and concern the transitional capacity assurance mechanism, are set out in more detail in section 16.3.2 below.

16.2.3 Generation adequacy In 2012, the wholesale electricity market showed signs of increasing maturity. The substantial new capacity that had entered the system in 2010–2012 intensified competition for mid and peak demand; it is notable that the total installed capacity of gasfired plants has reached that of lignite plants. The most obvious result of this kind of competition was the almost total replacement of electricity generation from aged and expensive oil-fired units by generation from gas-fired units. PPC’s share in the generation sector was substantially reduced to 79.5 per cent, in the Greek interconnected system, by taking into account only conventional technologies, and to 63.5 per cent if RES capacity is also included. In addition, the country’s continuing economic recession has reversed the demand growth trend that existed prior to the crisis and, together with the increased penetration of RES, this has resulted in reduced load factors for conventional power plants, a phenomenon that has severely challenged the financial viability of new power plants in particular. More precisely, while in 2008 the total net demand in the system reached 56.3 TWh, presenting an increase of 1.11 per cent in comparison to 2007, during the period 2008–2012 electricity demand decreased by approximately 6 per cent. However, demand is expected to increase due to future higher growth rates and the interconnection of Cyclades islands and Crete to the mainland system.25 Meanwhile, in 2012, the RES production in the interconnected system has reached approximately the level of 5,6 TWh. Wholesale prices, already relatively low and not reflecting the full variable cost of the marginal plants,26 were further suppressed by the significant amount of 21

See section 5.3.1, n 17 and the accompanying text. See RAE’s proposals (n 17). 23 See below, section 16.3.2, regarding Decisions 338/2013 and 339/2013 (n 9) amending the provisions governing the transitional CAM, currently in force. 24 RAE, Announcement of 18 July 2013 regarding the adoption of a set of transitional regulatory measures, available at http://www.rae.gr/site/categories_new/about_rae/factsheets/2013/major/18072013.csp, accessed 1 February 2015. 25 ENTSO-E, Scenario Outlook and Adequacy Forecast 2014–2030, June 2014 (ENTSO-E’s Report) p 77. 26 In Greece, a capacity market has already been in operation since 2006. Thus, the Greek market does not reflect the energy-only model and it is not expected from the wholesale energy prices to reflect the full production cost of the marginal plants but at least the variable cost of such plants. 22

292

Greece

electricity which entered the system on a priority basis. This includes mandatory hydro, plants’ minimum operational levels, and renewables.27 According to ENTSO’s Scenario Outlook and Adequacy Forecast 2014–2030, Greece has currently sufficient generation capacity. However, as figure 6.15 of ENTSO-E’s Report illustrates, the lack of new investment in power plants coupled with a possible increase in demand, results in declining generation adequacy in the coming years.28 ENTSO-E’s Report points out that although the Greek interconnected system has an installed capacity of ca 17.5 GW, which is significantly higher than the current and forecasted peak demand of 10–12 GW, the generation adequacy is only marginally positive until 2016. The reason is that generation adequacy depends on the available capacity rather than the installed one. As Table 16.1 shows, a large part of the installed capacity in Greece is composed of intermittent RES which make a limited and unstable contribution to the available capacity. Additionally, the aged fleet of lignite-fired units results in relatively high non-availability rates for the lignite capacity. Table 16.1 below shows the installed capacity in Greece’s interconnected system. As far as the available capacity in Greece is concerned, the results of ENTSO-E’s Report are illustrative. ENTSO-E uses a number of indicators in a ‘basic’ as well as in an ‘ambitious’ scenario to make a forecast on the projected capacity availability with regard to three points of reference: 2013, 2016, and 2020. The methodology followed corresponds to the one adopted by ENTSO-E in its annual forecasts. According to ENSTO-E’s findings, the Greek electricity system has adequate generation capacity for 2013. For 2016, the available capacity is considered satisfactory. However, it will be necessary to import energy from neighbouring countries to cover both winter and summer night peak demand. In particular, for the 2020, these imports are expected to reach 370 MW–460 MW. What is crucial is that under the scenario that four or more units are withdrawn, the system’s adequacy becomes insufficient and highly dependent on imports, where the total interconnection capacity will have to be used.29 Subsequently according to this analysis, the volume of available capacity is expected to remain limited in the Greek interconnected system. However, the operation of the electricity market should guarantee the viability of the available capacity. As long as an Table 16.1 Total installed capacity in Greece, 2013 SOURCES

MW

%

Thermal units Hydroelectricity units RES and cogeneration units TOTAL INSTALLED CAPACITY

10,238.5 3,017.7 4,169.81 17,426.01

58.8 17.3 23.9 100

Source: Author’s own table based on data provided by ADMIE and LAGIE.

27 According to Art 16(2) Directive 2009/28/EC of 23 April 2009, on the promotion of the use of energy from renewable sources and amending and subsequently repealing Directives 2001/77/EC and 2003/30/EC (2009 RES Directive). See also RAE, Annual Report to the European Commission (2013) p 7. 28 ENTSO-E’s Report (n 25) p 77. 29 ADMIE, Generation Adequacy Study for the period 2013–2020 (October 2013), available at http:// www.admie.gr/perissoteres-anakoinoseis/anakoinosi/article/1323/, accessed 1 February 2015, pp 31–51.

16.3 Capacity mechanism

293

energy-only market in Greece could not have guaranteed the necessary revenue streams for conventional power plants, a capacity mechanism appears prima facie to be necessary so that the units which contribute to the adequacy and flexibility of the system can benefit.

16.3 Capacity mechanism In 2002, when the market liberalization process was still under way, RAE proposed, for the first time, the implementation of a capacity mechanism for security of supply reasons. In its 2003 Annual Report,30 it highlighted that ‘taking into consideration the immaturity of the Greek electricity market and the necessity for new investment in generation capacity in the country, not only for the development of competition but also for security of supply, the adoption of a capacity mechanism in Greece would be highly beneficial.’ As briefly mentioned earlier, in section 16.2.1, there are two types of capacity mechanisms historically adopted in Greece. The first type is capacity obligation called Capacity Assurance Mechanism (CAM). This mechanism was adopted in 2005, but has almost never been implemented in practice.31 Instead, Greece adopted and implemented another type of capacity mechanism, a less complex and temporary mechanism based on capacity payments called Transitional Capacity Assurance Mechanism (transitional CAM).

16.3.1 Capacity Assurance Mechanism (CAM) The Law 3175/200332 created a legislative framework to introduce a market-based capacity mechanism delegating its implementation to RAE by means of executing acts. The Capacity Assurance Mechanism (CAM) was established for the first time in the 2005 Grid Code33 (as later amended in 2012)34 and currently is still in force. The CAM is a capacity obligation scheme based on certificates and contracts for differences. According to the 2012 amendment to the Grid Code, electricity producers are obliged to issue annual Capacity Availability Tickets (CATs), which incorporate a declaration that the production unit is going to technically maintain a certain level of available capacity in the future.35 However, certain units are exempt from this obligation, such as emergency services units, ancillary services units, renewable energy units, heat and power cogeneration units, and units connected to the distribution grid.36 This means that in reality only conventional and hydroelectric energy units are eligible to 30 RAE, 2003 Annual Report, available at http://www.rae.gr/site/file/system/docs/ActionReports/ report2003_2004, accessed 1 February 2015. 31 Actually only one contract has been concluded between PPC and a private electricity producer in 2007. 32 The Law 3175/2003 on the exploitation of geothermal resources, district heating, and other provisions, GG A 207, 29 August 2003, Art 23, chapter C. 33 Grid Code (n 20). 34 RAE, Decision 56/2012, GG B 104/2012 (the Greek Exchange Code); RAE, Decision 57/2012, GG B 103/2012 (the Greek Transmission System Operation Code). 35 36 Grid Code (n 20) Art 192(1)–(2). Grid Code (n 20) Art 191(2).

294

Greece

participate in the capacity mechanism, under the condition that they are installed on Greek territory. CATs are submitted to the CAT Register kept by the TSO. In legal terms, CATs are actually offers to the load representatives (who are mainly electricity suppliers) for the conclusion of Capacity Availability Contracts (CACs).37 Load representatives are obliged by the Grid Code to prove that they will be able to cover their supply obligations (plus a security margin) at all times, by presenting sufficient guarantees. In order to comply with this obligation, load representatives enter into CACs with electricity producers.38 These contracts have a standard template created by the Grid Code and they do not contain any financial agreements. Instead, the price is determined by the market and is the object of a separate agreement between the parties. As already mentioned, this mechanism has almost never been implemented in practice. Instead, a transitional Capacity Assurance Mechanism is being applied, which is analysed next.

16.3.2 Transitional Capacity Assurance Mechanism (Transitional CAM) In addition to the provisions mentioned earlier, a parallel, transitional capacity mechanism was considered necessary, given the immature nature of the recently deregulated energy market. The provisions establishing the Transitional Capacity Assurance Mechanism (transitional CAM) were added to the Grid Code in 2005. The transitional CAM is based on a capacity payment which is flat on all capacities and set administratively by the energy regulator. Initially, the capacity payments were given to all power plants regardless of their actual operation (but under the provision of being available at all times) and set at 35,000 €/MW-year. Despite its name, the transitional CAM is still in force. According to the provisions of the Grid Code, it will be retained as long as it is considered necessary and subject to annual evaluation by the energy regulator.39 Following its last evaluation, RAE extended the operation of the mechanism for one more year, until 30 September 2014.40 This prolongation is likely to be further extended. In the context of the transitional CAM, load representatives buy CATs from electricity producers at a regulated price (capacity payment). All capacity transactions are settled through an account run by the TSO. The role of the TSO in these transactions is to calculate and collect the sums owed by the load representatives and then credit the sums to each electricity producer, in proportion to the actual capacity they have offered. The provisions which govern the transitional CAM were recently amended by two regulatory decisions adopted by the energy regulator in June 2013. The energy regulator proceeded in adopting these two decisions as a temporary measure to address some of the distortions currently prevailing in the Greek market, until more permanent 37

38 Grid Code (n 20) Art 194(1). Grid Code (n 20) Art 188. Grid Code (n 20) Art 288 ff. Art 288 stipulates that the transitional provisions pertaining to capacity assurance according to Arts 288–291 of the Code are effective up to Reliability Year 2013–2014. 40 RAE, Decisions 338/2013 and 339/2013 (n 9). 39

16.3 Capacity mechanism

295

solutions could be implemented to guarantee undistorted competition conditions in the Greek market.41 The relevant market distortions are caused by two main factors. First, PPC, the incumbent energy company, still dominates the market, inter alia due to its exclusive right to use certain energy sources, such as lignite and most of the hydroelectric capacity available in the country. Secondly, PPC is still state-controlled and vertically integrated, a fact that often facilitates abuses of market power.42 According to the energy regulator, PPC’s share in the supply market was 98.2 per cent in 2004, 97 per cent in 2005, and 99.6 per cent in 2006, which shows its ability to exercise market power towards its suppliers.43 In addition, in 2012 four privately owned companies which were active in electricity supply abruptly exited the market. At the end of 2012 there were only six active suppliers (PPC plus five independent suppliers), and PPC enjoyed 99.9 per cent of the market share in the retail market.44 Currently, according to official data presented in August 2014 the market share of PPC in the retail market amounts to 96.37 per cent. To mitigate market distortions, the energy regulator has introduced the following amendments to the transitional mechanism. First, it introduced the possibility to (a) extend the application of the transitional CAM for one year (until September 2014), (b) RAE holds the option to remove from the CAT Register those plants which have exhausted their technical and economic life and are financially and technically inefficient compared to similar units being dispatched. Secondly, it introduced the possibility of a double CAT for power plants of modern technology, necessary to ensure generation adequacy during the period of market restructuring. As a result, nine plants which belong to PPC were removed from the CAT Register and eleven modern power plants, belonging both to PPC and independent producers, were registered for the double CAT.45 Lastly, capacity payment received by the producers in exchange for certificates was set at the level of 56,000 €/MW-year. It is crucial to stress that the two decisions46 of the energy regulator create regulatory measures which are supposed to be extraordinary and time restricted, as they will remain in force in principle until September 2014. Their aim is to ensure the viability of the existing production units in the current adverse economic conditions.47

16.3.3 The future of the capacity mechanism in Greece Given the previously mentioned characteristics of the Greek electricity market and EU’s general policy trends on capacity mechanisms, the energy regulator is currently 41

RAE, Decisions 338/2013 and 339/2013 (n 9). See also RAE, Announcement (n 24). See Decision 338/2013 (n 9) p 7. 43 See RAE, Annual Report 2006–2007, July 2007, p 11. 44 See RAE, National Report to the European Commission, December 2013, p 8. 45 According to the preamble of the Decision 339/2013 (n 9), units, which are of modern technology and bear the asymmetrical burden of the non-competitive structure of the electricity market, are allowed to register for the double payment since they will be most negatively affected in the absence of regulatory intervention (at p 7). 46 RAE, Decisions 338/2013 and 339/2013 (n 9). 47 RAE, Decisions 338/2013 and 339/2013 (n 9), preamble, which provides the objectives of the adopted measures. 42

296

Greece

exploring a revision of CAM, a proposed version of which recently went to public consultation.48 For the time being, this initial proposal sets out the main principles and a methodology for redesigning the existing CAM, in line with the Commission’s Generation Adequacy SWD.49 According to the RAE’s initial proposal,50 the current CAM (including the provisions on the permanent mechanism and on the transitional one) has to be rationalized, in accordance with the principle of proportionality and free market rules. In this direction, the future mechanism, consisting of a capacity payment scheme, will have to take into account that the sole aim of a capacity assurance mechanism is security of supply and the flexibility of the system whereas market failures51 should be addressed separately by other means. Additionally, the competent authorities have to take into consideration the economic output of every production unit as well as the extent to which every production unit contributes to the country’s security of supply. Under these circumstances, the future capacity mechanism should correspond to the system’s needs as these are calculated by ADMIE in the Generation Adequacy Study,52 whose criteria, calculation methodology, and requirements will also be subject to reevaluation by the TSO, taking into account the methodology adopted by ENTSO-E.53 Taking into consideration the Commission’s approach set out in its Generation Adequacy SWD, the new capacity mechanism will aim to achieve technology neutrality, decarbonization, and the participation of cross-border capacity through interconnections.54 This means that all types of electricity production technology, including renewables as well as units located abroad, would have the possibility to participate.55 As far as the time schedule is concerned, the capacity mechanism should be regularly revised in line with a clear roadmap for addressing underlying market and regulatory failures; thus, it should be designed to deliver a zero price when there is sufficient capacity available. The lead time for a capacity mechanism should correspond to the time needed to construct new power plants, that is, two to four years.56 Generation adequacy must be more clearly defined and its assessment must take into consideration all possible factors such as: interruptions and maintenance periods during which some generating units remain unavailable, a 48 RAE, Announcement of 27 July 2014. The consultation period ended on 17 September 2014. On the basis of information and opinions collected in the consultation process, RAE drafts its final proposal and submits to the Government within a few months. At the time of writing this chapter (1 October 2014), no proposal has been submitted yet. 49 Commission staff working document, Generation Adequacy in the internal electricity market— guidance on public interventions, 5 November 2013 (Generation Adequacy SWD) p 12. The EU policy on capacity mechanisms is discussed in chapter 1. 50 RAE, Proposal for the reorganisation of the Capacity Assurance Mechanism in the interconnected system, 29 July 2014, pp 1–18, available at http://www.rae.gr/site/file/system/docs/misc1/20102011/ 29071401, accessed 1 February 2015. 51 According to RAE, the current market structure fails to provide long term financial incentives for necessary investments in energy infrastructure (RAE, Proposal (n 50) p 3). Price caps or administratively set retail prices constitute such a negative example. 52 ADMIE (n 29). 53 ENTSO-E, Target Methodology for Adequacy Assessment (Updated Version after Consultation, 14 October 2014). 54 55 Generation Adequacy SWD (n 49) pp 26–31. RAE, Proposal (n 50) p 12. 56 RAE, Proposal (n 50) p 28.

16.4 European dimension

297

unit’s flexibility and its ability to respond to the customer’s profile; the primary energy sources availability, and the strategic reserves. Furthermore, interconnections are not at the moment considered eligible to participate in the proposed mechanism as their availability is considered highly uncertain, mainly due to Greece’s geographical position, despite the fact that RAE is reviewing their participation options.57

16.4 European dimension The design and implementation of capacity mechanisms usually raise questions regarding their compatibility with EU law and, more specifically, state aid rules and free movement provisions. As a result, a debate has arisen concerning the Greek capacity mechanism and its compatibility with EU law.

16.4.1 Compatibility of the transitional CAM with EU state aid rules Concerns have been raised both by PPC before the Greek Council of State as well by a complainant before the Commission that the two decisions of the energy regulator amending the transitional CAM58 constitute state aid within the meaning of Article 107(1) TFEU. As a result, the legality of these decisions has been challenged before the Supreme Administrative Court in Greece and a complaint has been filed with the Commission alleging that the two decisions constitute illegal state aid. At the time of writing this chapter,59 no formal investigation was opened by the Commission and the Council of State has not yet dealt with the case. Notwithstanding this fact, it would be perhaps useful to highlight some of the counterarguments posed as regards the classification of the two RAE’s decisions as state aid.60 The EEAG 2014–202061 recognizes the necessity to compensate generation capacity. In the Commission’s view, with the increasing share of renewable energy sources, electricity generation is in many Member States shifting from a relatively stable and continuous supply towards a system with more numerous and small scale supply of variable sources. This shift raises new challenges for ensuring generation adequacy. [ . . . ] As a result, some member states consider introducing measures to ensure generation adequacy, typically by granting generators support for the mere availability of generation capacity.62

57

RAE, Proposal (n 50) p 12. 59 RAE, Decisions 338/2013 and 339/2013 (n 9). 1 October 2014. 60 The author does not intend to formulate here a critical evaluation of the expressed counterarguments and express his own scientific opinion on the issues raised since this would presuppose a thorough analysis of all relevant techno-economic parameters of the issue, something that clearly lies out of the scope of the present analysis in this chapter. 61 Communication from the Commission, Guidelines on State aid for environmental protection and energy 2014–2020 [2014] OJ C 200/1–55 (EEAG 2014–2020). 62 EEAG 2014–2020 (n 61) paras 217 and 219. 58

298

Greece

A measure of public support is considered state aid only if the conditions defined in Article 107(1) TFEU are all satisfied.63 In the context of the transitional CAM, the main concern relates to the criterion of economic advantage.64 Namely, the level of capacity payment is regulated by law, instead of being determined by free function of the market, ie supply and demand. This regulated capacity payment might confer an economic advantage in the sense of Article 107(1) TFEU. In addition, the capacity payment is flat for all capacities and applies to all existing units, despite the fact that some units are still economically viable and would remain in the market regardless of state support. If the level of capacity payment wasn’t regulated, but determined by the market, opponents argue, the compensation for available capacity would at least have been differentiated among participants in the CAM and would not be uniform, as is currently the case. In response to this, one could raise the following counterargument. Provided that capacity payments cover power plants’ fixed costs only, participants in the transitional CAM are simply remunerated for providing public service (security of supply). Therefore, provided that there is no overcompensation, capacity payments should confer no economic advantage. This argument, however, requires an in-depth economic analysis of the annual fixed cost for the operation of a plant, which would show whether this cost exceeds or is at least equal to the capacity payment, which is currently set at 56,000 €/ MW. Assuming that this is the case (no overcompensation), it could be further argued that the transitional CAM compensates power plants for providing a PSO and therefore does not fall under the concept of state aid. However, in order to be considered as compensation for the PSO and not state aid, the transitional CAM must fulfill other criteria of assessment applied by the Commission and the European Courts.65 Regarding the criterion of state resources and state control,66 electricity producers in the transitional CAM are remunerated directly by load representatives via a special CAM account which includes individual accounts, separate for each participant. The CAM account is administered by the TSO, in which the Greek State holds a controlling stake (51 per cent) and can therefore be regarded as a public body. The TSO’s role is limited to collecting the revenues from the load representatives and delivering them to the beneficiaries. In other words, it acts as an intermediary in the financial transactions between producers and suppliers. Therefore, the assessment of CAM under state aid rules also requires a careful analysis of the mechanism’s link with the Greek State in the light of the case law of the CJEU.67 63

See section 9.3 for a discussion. A measure is classified as state aid only if, among others, it constitutes an economic advantage conferred on the undertaking(s), which could not have been obtained under normal market conditions. For further explanation, go to section 9.3.2. 65 For instance, the recipient undertaking must actually have a PSO to discharge and the obligations must be clearly defined. In addition, the parameters on the basis of which the compensation is calculated has to be established in advance. See Case C-280/00 Altmark Trans [2003] ECR I-07747. See also Antonis Metaxas, ‘Commentary on Articles 101–109 TFEU’ in Christianos (ed), Treaty on the Functioning of the European Union, A Legal Commentary (Nomiki Vivliothiki, 2012) pp 560–613. 66 To be considered state aid, a measure must involve the use of state resources and must be imputable to the state (see section 9.3.4 above). 67 Case C-482/99 France v Commission [2002] ECR I-4397 and C-72-73/91 Sloman Neptun [1993] ECR I-887. 64

16.4 European dimension

299

Regarding the criterion of selectivity,68 it needs to be examined whether certain producers which are in a similar factual and legal situation are treated differently, in which case the different treatment would have to be objectively justified by the nature or the general scheme of the system, so that the criterion of selectivity is not fulfilled.69 In the context of the transitional CAM, it is important to safeguard that all units are eligible to participate in the mechanism without distinction, as long as they contribute to ensuring the long-term generation adequacy. However, following the recent amendments to the transitional CAM, some units have been registered for a double CAT (modern power plants) and other have been excluded from participating in the mechanism (inefficient plants removed from the CAT Register).70 This made the opponents of the capacity mechanism claim that these amendments allow certain producers to receive double payments, while excluding other producers from participating in the mechanism, and thus from receiving any compensation at all. With regard to this argument, it must be pointed out that as long as different treatment of undertakings is objectively justified by the nature or the general scheme of the system, then the criterion of selectivity is not fulfilled. It must be stressed that only old and inefficient plants, which could not contribute to the country’s long-term security of supply, have been excluded from the transitional CAM. Provided that the selection of these plants was based on objective technical criteria of plant efficiency and flexibility, their exclusion can be regarded as justified by the ‘internal logic’71 of the system for which the transitional CAM was created. Lastly, concerns have been raised that the existing transitional CAM has not been notified to the Commission, in violation of the procedural obligation stipulated in Article 108(3) TFEU. Notwithstanding the previous discussion, Greece is planning to implement a new capacity mechanism compatible with the EU Target Model. Its design is currently under discussion. In that respect, the Greek government intends72 to send to the Commission a draft plan setting out the general framework of the new capacity mechanism.

16.4.2 Compatibility of the CAM with the free movement of goods rules Any state intervention to ensure generation adequacy must also be in line with the free movement of goods rules73 (in particular, Article 34 TFEU), so that it does not distort cross-border trade. In that respect, the combined reading of Article 4 and Article 191 of the Grid Code suggest that foreign producers are not able to participate in the Greek CAM (both permanent and transitional one) due to procedural barriers. Namely, CATs can only be issued by generating units registered in the Generation Unit 68 State aid control only applies if a state measure is selective, ie limited to certain undertakings or certain goods (see above, section 9.3). 69 C-143/99 Adria-Wien Pipeline and Wietersdorfer & Peggauer Zementwerke [2001] ECR I-8365, para 41. 70 RAE, Decision 339/2013 (n 9). See above, section 16.3.2. 71 Case C-75/97 Belgium v Commission (Maribel) [1999] ECR I-3671, para 18. 72 Energypress, article of 29 April 2014, available at http://www.energypress.gr/news/Ekplhxeis-sta-neaADI-mpainoyn-APE-alla-kai-zhthsh-Ta-krithria2, accessed 1 February 2015. 73 For an in-depth discussion, see chapter 11.

300

Greece

Register,74 which, in turn, is only open to plants located in Greece.75, 76 Similar procedural barriers can be observed in other capacity mechanisms currently implemented in Europe.77 However, even if a national measure impedes cross-border trade, it can still be justified on grounds of public security, as listed in Article 36 TFEU. The EU courts have recognized that the objective of security of supply can be a justification for a Treaty exemption, provided that the national measure pursuing this objective is appropriate and proportional.78

16.4.3 Compatibility of the CAM with the sector-specific provisions This section discusses the compatibility of the CAM with EU sector-specific provisions, in particular the Security of Supply Directive and the 2009 Electricity Directive.79 While the latter has been transposed into national law by Law 4001/2011, the Security of Supply Directive has not yet been fully implemented but no infringement proceedings have been opened.80 Article 5 of the Security of Supply Directive allows Member States to take appropriate measures to ensure generation adequacy. This includes provisions facilitating new generation capacity81 as well as tendering procedures or any procedures equivalent in terms of transparency and non-discrimination.82 This means that if a Member State contemplates implementing a capacity mechanism, it needs to be transparent and nondiscriminatory. Further, in the interest of security of supply, Article 8 of the 2009 Electricity Directive obliges Member States to provide for new capacity by means of a tender or equally transparent and non-discriminatory procedure. In that respect, the two recent amendments to the transitional CAM have been criticized for violating the principle of non-discrimination. In particular, it was claimed that excluding certain units from receiving the capacity payment results in discrimination between electricity producers. To the contrary, the energy regulator has argued that it removed certain plants from 74

Grid Code (n 20) Arts 4 and 191. The Greek mainland system and islands connected to the mainland system (the interconnected system, see section 16.2.1). 76 Grid Code (n 20) Art 4. 77 See capacity mechanisms in France (chapter 14) and the UK (chapter 22). 78 See, for instance, Case 72/83 Campus Oil v Minister of Industry and Energy [1984] ECR 2727 (Campus Oil) and Case C-398/98 Commission v Greece (Greek Oil Supplies II) [2001] ECR I-7915. See also discussion on these cases in section 11.5.4.1 above. 79 Directive 2005/89/EC of the European Parliament and of the Council of 18 January 2006 concerning measures to safeguard security of electricity supply and infrastructure investment [2006] OJ L 33/22 (Security of Supply Directive) and Directive 2009/72/EC of the European Parliament and of the Council of 13 July 2009 concerning common rules for the internal market in electricity and repealing Directive 2003/54/EC [2009] OJ L 211/55 (2009 Electricity Directive). 80 At the time of writing this chapter (1 October 2014). Regarding the implementation of the 2009 Electricity Directive (n 79), national acts have been adopted consisting only in the formation of interministerial committees which are assigned to draft the necessary legal documents to be adopted. See Ministerial Decision Y` ˜13/.7.1/13624 GG B 1058, 4 August 2014. 81 Security of Supply Directive (n 79) Art 5(a). 82 Security of Supply Directive (n 79) Art 5(f). 75

16.5 Conclusion

301

the CAT register for valid reasons, as these plants have exhausted their lifespan and are substantially less efficient than other dispatched units.83

16.5 Conclusion The analysis in this chapter gives ground to questions on the incompatibility of certain aspects of the Greek capacity mechanism with provisions of EU law. In this framework and within the scope of the restructuring of the Greek energy market, the aim of the energy regulator and the legislator is to introduce a revised, permanent mechanism which will achieve its goal in offering the necessary incentives to electricity generators without introducing unlawful state intervention. Besides, the energy regulator has directly expressed the need that the future capacity mechanism is in line with the respective European legal framework. The need to restructure the Greek capacity mechanism is also an obligation arising from Greece’s commitments with the Troika and the Memorandum of Understanding signed with its European and international counterparts for the financing of its debt. However, given the fact that the initial proposal of the energy regulator is still under consultation, it is obvious that the initial time schedule agreed with the Troika has been missed. Instead, the final proposal on the future mechanism, which will be submitted to the Commission for its evaluation according to the new guidelines, will not be ready until the beginning of 2015.

83 RAE, Decision 339/2013 (n 9), and also the Grid Code (n 20) Art 191(6), as amended, where the specific reasoning is set out.

17 Italy Francesco Maria Salerno1

17.1 Introduction Italy is not an energy-only market: since 2003 eligible generators receive a capacity payment. In spite of being temporary, the rules enacted in 2003 are still applicable. After a long process, in September 2013 new rules concerning the principles of the ‘final’ capacity mechanism were approved. However, in December 2013 Parliament passed a law requiring further changes affecting both the temporary and the final system; in June 2014 these changes had yet to be adopted. As a consequence, this chapter provides a snapshot of the situation at the time of writing.2 The chapter is structured as follows. Section 17.2 provides background information on the market and the regulatory framework. Section 17.3 provides information on the existing and the planned capacity mechanism. Finally, section 17.4 touches on the European dimension.

17.2 Setting the scene 17.2.1 Market characteristics In Italy, electricity producers and importers sell electricity to suppliers, traders, and large industrial customers in the market pool (mercato elettrico) through a bidding process. The market operator (Gestore Mercati Energetici, GME) arranges bids in a merit order.3 Electricity sellers earn their revenues on the basis of marginal pricing. Due to network constraints, the wholesale electricity market in Italy is divided into geographical and virtual zones.4 Producers’ revenue is set at the zone level, while buyers pay a price that is the weighted average of the zonal prices (prezzo unico nazionale, PUN). There are currently six zones, as shown in Figure 17.1 below.5

1 The author wishes to thank Alice Setari for her invaluable help in research, drafting, and editing the chapter. 2 3 June 2014. See section 1.1. 4 Certain zones are ‘virtual’ because they are a regulatory artifact, whose purpose is to give separate evidence to the import/export from/to Italian electrical borders. A ‘macro-zone’ is an ‘aggregation of geographical and/or virtual zones that is conventionally defined for the production of statistical market indexes. A macro-zone has a low frequency of market splitting and a homogeneous trend of selling prices’, according to the definition provided by GME, available at http://www.mercatoelettrico.org/en/Tools/ Glossario.aspx#M, accessed 1 February 2015. 5 As of 1 January 2009 there are four macro-zones: MzNord (the northern area and Monfalcone), MzSicilia (Sicily and Priolo zones), MzSardegna (the Sardinian area), and MzSud (the remaining areas).

17.2 Setting the scene

303

65,09 (+8,1%)

62,36 (+6,1%) 62,36 (+6,2%)

61,63 (+4,9%) 62,40 (+6,2%)

107,42 (–0,5%) Scarto % dal PUN +5%

Figure 17.1 Price zones in the Italian electricity market Source: Author’s own illustration.

Wholesale prices in 2014 were on average 52 €/MWh for Italian day-ahead base load power (a decrease compared to the previous years (63 €/MWh in 2013 and 75 €/MWh in 2012)).6 Nevertheless, in a recent report on the EU energy markets, the Commission observes that ‘the average price of electricity [is] above the rest of the Europe due to the generation park composition by far led by combined cycles gas fired plants.’7 The figures below provide information on electricity producers and their market shares in terms of production (Figure 17.2) and coverage of demand (Figure 17.3). Figure 17.4 illustrates the Italian generation mix, with a 68 per cent share of thermal generation, and 32 per cent share of RES. Another important feature of the Italian electricity market is the percentage of electricity imported into the country: in 2012 imports accounted for approximately 13 per cent of demand.8 This is a long-term trend. Italy has historically imported electricity and has had long-term contracts with producers in France and Switzerland dating well before the market first opened in 1999. As noted by the Commission in its report on the EU energy markets, Italy remains one of the best interconnected European countries.9 6

Source: GME. European Commission, EU Energy Markets in 2014, (adapted version of the Commission Staff Working Documents SWD (2014) 310 final and SWD (2014) 311 final) 13 October 2014, available at http://ec.europa.eu/energy/sites/ener/files/documents/2014_energy_market_en_0.pdf, accessed 1 February 2015, p 87. 8 AEEG, Annual Report 2013, Vol I, p 53: ‘In 2012, national net production covered more than 87.5% of requirements, in line with the previous period, with net imports (43 TWh) which helped to cover the remainder of demand. Exports are the only voice in countertrend, with an increase of 27.6%.’ (Author’s own translation.) 9 European Commission (n 7) p 87. 7

304

Italy

Enel Eni Edison E.On Edipower Gdf Suez A2A Tirreno Power Erg Iren Sorgenia Axpo Group Saras Altri produttori 0%

25.4% 26.2% 9.5% 9.3%

7.2% 8.3% 4.4% 5.2% 3.9% 4.8% 3.6% 3.1% 3.2% 3.5% 3.1% 3.7% 2.9% 2.5% 2.2% 2.0% 1.9% 2.0% 1.7% 1.5% 1.6% 1.5%

26.3%

5%

10%

15%

20%

2012

25%

29.6%

30%

35%

2011

Figure 17.2 Contribution of major groups to domestic gross production: comparison 2011–2012 Source: AEEG, Annual Report 2013, vol I, p 57.

Enel Eni Edison E.On Edipower Tirreno Power A2A Gdf Suez Erg Iren Sorgenia Axpo Group Saras Altri operatori 0%

26.1% 10.4% 7.9% 4.7% 4.1% 3.2% 3.4% 3.9% 3.2% 2.4% 2.1% 1.9% 1.7% 25.0% 5%

10%

15%

20%

25%

30%

Figure 17.3 Contribution of major groups to production of electricity for consumption in 2012 Source: AEEG, Annual Report 2013, vol I, p 59.

17.2.2 Regulatory framework The Ministry of Economic Development and the Autorità per l’Energia Elettrica e il Gas (AEEG, energy regulator) share responsibility for the overall supervision and regulation of the Italian energy sector, which comprises both electricity and gas. In general, the Ministry of Economic Development is responsible for establishing strategic guidelines for the energy sector and for ensuring the safety and economic soundness of the electricity and gas sectors. AEEG is responsible for setting and adjusting tariffs on the basis of general criteria established by law, advising the Ministry of Economic Development on the structuring and administration of licensing and authorization regimes

17.2 Setting the scene

305 Thermal Power Generation (68%)

2% 6%

Solids 4%

Natural Gas

16%

Petroleum products

5%

Other Production from Renewable Sources (32%)

15%

5%

Hydroelectic 44%

Wind Power Photovoltaic Geothermal

3%

Biomass and Waste

Figure 17.4 2012 net production by source Source: Author’s own illustration based on AEEG, Annual Report 2013.

for the energy sector, ensuring the quality of services provided to customers; overseeing the separation of utility companies into distinct units for accounting and management purposes, promoting competition, and otherwise protecting the interests of consumers, including the authority to mediate disputes between utilities and consumers, and to impose sanctions for violations of regulations. The main objective of the government’s energy policy is the pursuit of security and adequacy in the electricity sector.10 This same objective is also referred to in documents by AEEG.11 The regulatory framework for the Italian electricity sector has changed significantly in recent years following the implementation of the EU Electricity Directives.12 The Legislative Decree No 379/2003 of 19 December 2003, Article 1: ‘The remuneration system for the availability of electricity production, regulated by this decree, ensures the achievement and maintenance of the adequacy of production capacity in order to ensure coverage of domestic demand with the necessary reserve margins.’ (Author’s own translation.) See also ACER, Report on capacity remuneration mechanisms and the internal market for electricity, 30 July 2013 (ACER’s Report) p 9. 11 AEEG, Annual Report 2013, Vol II, p 37: ‘Finally, with regard to the integration in the electricity market of a growing share of renewable energy sources, the Authority noted that the initiative to identify the critical areas with a high concentration of non-programmable renewable sources and to limit the capacity eligible for incentives in these areas is correct and necessary. In order to ensure maximum effectiveness, however, such an initiative should be coordinated in terms of deadlines, time horizons, and quantitative goals with forecasting and planning carried out by Terna in the new market of production capacity, outlined by Legislative Decree No 379/2003 (n 10), and by Resolution ARG/elt 98/11 of 21 July 2011 (capacity market). Without a coherent development plan for non-programmable generation capacity, you risk compromising in the long term the coordinated development of generation capacity and of the National Transmission Network.’ (Author’s own translation.) 12 Directive 96/92/EC of the European Parliament and of the Council of 19 December 1996 concerning common rules for the internal market in electricity [1997] OJ L 27 (1996 Electricity Directive). Directive 2003/54/EC of the European Parliament and of the Council of 26 June 2003 concerning common rules for the internal market in electricity and repealing Directive 96/92/EC [2003] OJ L 176/37 (2003 Electricity Directive). Directive 2009/72/EC of the European Parliament and of the Council of 13 July 2009 concerning common rules for the internal market in electricity and repealing Directive 2003/54/EC [2009] OJ L 211/55 (2009 Electricity Directive). 10

306

Italy

foundation of the current system is the so-called ‘Bersani Decree’13 which put in place the 1996 EU Electricity Directive.14 The Bersani Decree entered into force on 1 April 1999 and began the liberalization of the electricity sector through the separation of generation, transmission, and distribution activities. It gradually introduced free competition in power generation and sales to consumers meeting certain consumption thresholds, while maintaining a regulated monopoly structure for power transmission, distribution and sales to the other customers. In 2003 the EU adopted the Second Electricity Directive15 to enable all consumers to freely choose their supplier by 2007, irrespective of consumption levels. Italy accordingly adopted implementing measures, the so called ‘Marzano Law’.16 Based on the Marzano Law, electricity transmission services became fully unbundled from production and supply activities. Responsibility for the management of the national transmission grid and the related assets was transferred from the Gestore della Rete di Trasmissione Nazionale (GRTN) to Terna, although the GRTN retained its other responsibilities.17 From 2 October 2006, GRTN was renamed Gestore dei Servizi Elettrici (GSE). GSE is mainly responsible for administering the incentive schemes for renewable energy sources. By the end of September 2011, Italy had fully transposed the Third Energy Package into national law.18

17.3 Capacity mechanism 17.3.1 The existing capacity mechanism Italy is not an energy-only market: since 2003, the energy-only market has been supplemented by a capacity mechanism. The Legislative Decree No 379/2003 created a mechanism administered by AEEG.19 More specifically, on the basis of the Legislative Decree 379/2003 (Article 5), pending the definition of the final regime, AEEG adopted a temporary system in 2004, known as the capacity payment system regulated by AEEG Decision 48/04.20 It is based on awarding an administered fee to those operators which make their capacity available to Terna. The temporary system will cease to exist with the entry into force of the final regime.21 The fee in the temporary system has two parts. First, eligible plants (selected based on their reliability) receive a basic remuneration determined in advance, based on 13

Legislative Decree No 79 of 16 March 1999 (Bersani Decree). 15 1996 Electricity Directive (n 12). 2003 Electricity Directive (n 12). Law No 239 of 23 August 2004 (Marzano Law). 17 This occurred on 1 November 2005 following an implementing decree enacted in May 2004. 18 See European Commission, Energy Markets in the European Union in 2011, SWD(2012)368, 15 November 2012, Country reports: Italy, p 1. 19 Decree No 379/2003 of 19 December 2003. This measure defined principles to be adhered to (competition, transparency, no discrimination, no market distortions) and set a timeline for the introduction of reform. 20 AEEG, Decision 48/04 of 27 March 2004. This measure provided the temporary regime to be applied until the entry into force of the final capacity mechanism. 21 See also Regulatory Commission for Electricity and Gas (CREG), Capacity remuneration mechanism, Study (F)121011-CDC-1182, 11 October 2012, p 11 (CREG’s study on capacity mechanisms). 14 16

17.3 Capacity mechanism

307

forecasts (not actual data) of supply and demand for each hour of the following day. Secondly, the eligible plants receive a further payment if the weighted average price on the day-ahead market is less than 20 per cent of a reference price determined by the AEEG and if the plant is located in an area with low prices (Enel, for example, does not receive this additional payment). The participation in the scheme is voluntary. But the plants which qualify must be available in the day-ahead market during peak periods, as well as on certain critical days of the year. In December 2013 Law No 147 required AEEG to propose changes to the transitory mechanism.22 AEEG concluded its public consultation on 3 June 2014.23 The adoption of the changes to temporary regime should take place before the end of 2014.

17.3.2 The final capacity mechanism In Italy the capacity problem is perceived as a market failure. According to AEEG Decision 98/11 of 21 July 2011, this market failure consists of information asymmetry: key information for investment decisions is considered incomplete and it is distributed asymmetrically between generators and the network operator, Terna. In the absence of regulatory intervention, the electricity market reveals itself to be inefficient and ineffective at coordinating the investment choices by generators and Terna. AEEG Decision No 98/11 goes on to justify regulatory intervention based on a number of factors which in its view exacerbate this market failure.24 These factors include the following: a high level of investment risk, sunk costs, the short-term electricity demand inelasticity, the slow adjustment of production capacity at any sign of shortage or oversupply of electricity, and the high cost of electricity storage. According to Decision 98/11, these factors amplify the effects of the investment cycle in measures incomparable with other sectors. This causes extreme price volatility in spot markets for electricity and ancillary services. Coupled with information asymmetry, price volatility further increases the level of risk for investors. Decision 98/11 also mentions the existence of regulatory failures: the rapid and growing penetration of non-programmable renewable sources in electricity generation, and uncertainties caused by the continuous and unpredictable evolution of environmental legislation. Moreover, according to Decision 98/11 risk aversion of consumers makes it almost impossible to conclude long-term contracts. This failure distorts investment choices among the many technologies/sources of electricity generation because investors tend to choose technologies that determine most of the time the equilibrium price on the market, ie those whose variable costs are more related to the market price. However, this generator mix is not necessarily the optimal one from the point of view of the system as a whole. 22

Law No 147/13 of 27 December 2013, Art 1, para 153. AEEG, Consultation Document of 22 May 2014, No 234/2014 R/EEL, available at http://www. autorita.energia.it/it/docs/dc/14/234-14.jsp, accessed 1 February 2015. 24 AEEG, Decision 98/11 of 21 July 2011 (Considerando). See also AEEG, ‘Elettricità: in arrivo l’“assicurazione” fra produttori e consumatori per ridurre rischio-prezzi e promuovere investimenti’, Press release of 22 July 2011, available at http://www.autorita.energia.it/it/com_stampa/11/110722_capacity. htm, accessed 1 February 2015. 23

308

Italy

Energia Concorrente, an Italian trade association representing the interests of new comers has made its views known by agreeing with the proposal to introduce a marketbased capacity mechanism.25 Moreover, Energia Concorrente considers that a capacity mechanism appears to be necessary as the sharp increase of renewable sources (fuelled by a generous subsidy system) is reducing the contestable market (conventional power plants operate today in a substantially smaller competitive environment) and risks jeopardizing security of the electricity system and generation adequacy. According to AEEG, the temporary regime entailed costs for approximately €150 million in 2013.26 It is difficult to reliably predict the costs for the final capacity mechanism as it depends on future auctions, whose outcome can be influenced by several factors.

17.3.2.1 Key elements Under the proposed capacity mechanism,27 Terna has to define the generation adequacy objective at national and zonal level each year for each of the following ten years.28 On the basis of annual updates of the estimated capacity gap, Terna will decide if and how much capacity to auction. The price is set by the auction (lowest bid) and paid to all participants who are successful in the auction (for each MW of committed capacity, operators will receive an annual fee in €/MW). It is estimated that a development plan of at least four years will be necessary,29 therefore, assuming that they take place in 2015, first auctions will bind capacity first in the year 2019. The contract will bind the counterpart to make production capacity available for a period of not less than three years. The capacity mechanism therefore sets out a system of permanent revisions. The selected participants must sell the capacity in the market. However, the participant must make the following payments, depending on the difference between the exercise price, equal to the standard variable production cost of a specific technology Energia Concorrente gathers five founding companies, Axpo Italia, GDF Suez Energia Italia, Repower, Sorgenia, and Tirreno Power. See the Response of Energia Concorrente to the Commission’s consultation on generation adequacy (discussed in section 1.4.5) downloadable from the Commission’s website at http:// ec.europa.eu/energy/en/consultations/consultation-generation-adequacy-capacity-mechanisms-andinternal-market-electricity, accessed 1 February 2015, under non-registered organizations, p 5. 26 See AEEG, Consultation Document (n 23) p 12. 27 The key regulations include: (a) AEEG, Decision 98/11 (n 24) setting out the criteria and conditions for Terna to develop its detailed proposal for the capacity mechanism (Schema di proposta di disciplina del nuovo sistema di remunerazione della capacità produttiva di energia elettrica, available at http://www.terna. it/LinkClick.aspx?fileticket=104543, accessed 1 February 2015, ‘Framework Discipline’), (b) AEEG, Decision 482/2012, 15 November 2012, which was the first positive check of Terna’s Framework Discipline. Terna put the Framework Discipline to public consultation, which ended on 15 February 2013. Terna then delivered to AEEG its new proposal taking into account the results from the consultation, (c) AEEG, Decision 375/2013, 5 September 2013, granting a final approval of Terna’s Framework Discipline. See AEEG, Annual Report 2013, 31 March 2013, Vol II, pp 47–8, available at http://www.autorita.energia.it/ allegati/relaz_ann/13/RAVolumeII_2013.pdf, accessed 1 February 2015. The Minister of Economic Development had to adopt a Decree to endorse the proposal. However, Law No 147 of 2013 requires the AEEG to make further changes to the proposal. On 3 June 2014 AEEG concluded its consultation process (AEEG, Consultation Document (n 23) p 12). The adoption of rules for the final capacity mechanism is expected by the end of 2014. 28 AEEG, Decision 98/11 (n 24) Art 5. 29 AEEG, Decision 98/11 (n 24) Art 6(2). 25

17.3 Capacity mechanism

309

Reference Price Δ Exercise Price

Figure 17.5 The proposed capacity mechanism based on reliability options: reference price and exercise price Source: Author’s own illustration.

selected by Terna,30 and the reference price, ie the price fixed on the day-ahead market (for the capacity sold on that market) or on the market for ancillary services (for the capacity sold for the ancillary services on that market).31 Figure 17.5 illustrates the relation between the reference price and the exercise price. Thus, the Italian final capacity mechanism in essence follows a ‘reliability option’ model.32 Terna’s Framework Discipline, a document which sets out the details of the proposed capacity mechanism,33 does not discuss the possibility that capacity located abroad could also take part in the mechanism. However, should generators connected to the network of another European TSO express interest in participating in the Italian capacity mechanism, Terna and the TSO in question can explore ways to make such a cross-border participation possible and propose necessary amendments to the Framework discipline in that respect.34 Likewise, there is no direct prohibition for Italian national power plants to participate in capacity mechanisms abroad.

17.3.2.2 Eligibility criteria Eligibility criteria to take part in the tender procedure include the following: (a) availability of programmable production capacity,35 (b) proof of having existing or new production capacity that refers to significant production units located on the national territory, in particular the amount must be greater than 10 megavolt ampere [MVA],36 and (c) capital requirements.

17.3.2.3 Enforcement powers The TSO can suspend a participant from the capacity mechanism if they do not respect either the provisions of the Framework discipline or their contract or the requirements 30

31 AEEG, Decision 98/11 (n 24) Art 9. AEEG, Decision 98/11 (n 24) Art 10. For more information on this model, go to section 1.2.3.4. 33 34 Terna, Framework Discipline (n 27). AEEG, Decision 375/2013 (n 27) p 10. 35 AEEG, Decision 98/11 (n 24) Art 10(7). 36 AEEG, Decision 375/2013 (n 27) p 10. As noted earlier in the text, depending on expression of interests from foreign generators, the Framework discipline can be amended to take into account also generators located outside the national territory. 32

310

Italy

of the guarantee fund or capital requirements. The TSO can revoke the suspension if the participant remedies the breaches. Once a participant has been suspended for six months, the TSO will exclude it from the market.37 In addition, the TSO will inform AEEG about violations and AEEG may impose sanctions.38 In particular: [Terna] reports the breaches of the obligations to [AEEG], which shall impose sanctions commensurate with the gravity of the violations, ranging from a minimum value of 25,000 €/MW and a maximum value of 50,000 €/MW on an annual basis [ . . . ]. The sanctions are proportionate to the actual periods of unavailability, during the year [of] the capacity remunerated [ . . . ]. In cases of greater severity and recurrence of the violations, [AEEG] may order the suspension of the remuneration in respect of non-compliant operators. [AEEG] sets, on a proposal by Terna, the technical criteria for the calculation of the unavailability of the power remunerated.39

17.4 European dimension 17.4.1 Acknowledging the EU context The national measures that set out the final capacity mechanism refer to the 2009 Electricity Directive.40 Moreover, in March 2013, the Italian Parliament approved the National Strategy for Energy for the coming years.41 In relation to the planned Italian capacity mechanism, it expressly referred to the guidance of the Commission (which was issued afterwards, in November 2013), in particular the importance that any measure must avoid creating obstacles in the internal market or take the form of illegal public aid.42 In a recent public speech, AEEG’s President, Guido Bortoni, expressed confidence in the EU-wide energy market by stating the following: Making projects is essential in the field of regulation, just as it is in the more general field of energy and environmental policy. It also becomes positive forward planning if it forms part of a responsible approach to regulation which does not allow us to imagine—not least out of respect for generations to come—a future world which is an easy dumping ground for problems which remain unsolved or are put off today. A responsible regulation gives no leeway to those who believe—from a divisive perspective—that it is, selfishly, only the present which should be regulated—better still if restricted to the domestic context. As if problems and solutions, including those in the field of energy, were not European in scope and significance. With regard to this latter point we may draw an effective lesson from the recent past. Do we really wish to provide individual, domestic responses, such as those chosen after the oil price shocks 37

Legislative Decree 379/2003 (n 19) Art 3, and Terna, Framework Discipline (n 27) section 1.6. Legislative Decree 379/2003 (n 19) Art 4. 39 40 Author’s own translation. 2009 Electricity Directive (n 12). 41 The Italian Energy Strategy has been enshrined in Ministerial Decree of 8 March 2013, available at http://www.sviluppoeconomico.gov.it/images/stories/normativa/decreto-8marzo2013-sen.pdf, accessed 1 February 2015. 42 Ministry of Economic Development, National Strategy for Energy, March 2013, available at http:// www.sviluppoeconomico.gov.it/images/stories/normativa/20130314_Strategia_Energetica_Nazionale.pdf, accessed 1 February 2015, p 103. 38

17.4 European dimension

311

of the 1970s, which have left us the legacy of a Europe with high explicit and implicit costs and clear energy and environmental imbalances in its various Member States? It may be said that today’s ‘Europe of energy’ struggles to provide valid solutions; yet recourse to a reckless energy jingoism is a losing path for everyone [ . . . ].43

However, the public documents make reference neither to neighbouring countries’ choices, nor the impact of the Italian mechanism on such countries.

17.4.2 Assessment of capacity mechanisms under EU law 17.4.2.1 The internal market dimension and the Italian experience One of the central themes emerging from the Commission’s November 2013 Communication44 and public statements is the concern that capacity payments may constitute a hindrance to the completion of the single energy market. For instance, according to ACER, capacity payments ‘may potentially distort cross-border trading or even act as a barrier to trade if they are designed without taking into account their cross-border impact or are implemented at national level without any coordination with neighbouring jurisdictions’.45 Indeed, interconnection capacity/imports and capacity mechanisms are often pitted against each other. For instance, the EEAG 2014–2020 require, as a condition to approve capacity payments as compatible aid, that capacity mechanisms do ‘not reduce incentives to invest in interconnection capacity.’46 The Italian case provides interesting insights on the interplay between interconnection capacity/imports and capacity payments. Italy has had a capacity mechanism since 2003 and yet during this time imports have constantly been significant (above 10 per cent of demand, the so-called ‘Barcelona target’). There are several explanations for this, including that: (a) the price differential between Italy and neighbouring countries remain large, even if capacity payments are available, (b) the remuneration under the capacity mechanism is limited. The space of this chapter does not allow for an in-depth analysis of these factors. Nevertheless, the fact remains that the Italian case proves that capacity payments do not necessarily lower incentives for imports. Moreover, Italy introduced the capacity mechanism after experiencing one of the worst black-outs in European history. One of the factors explaining the black-out was Italy’s dependence on imports. Thus, the co-existence of capacity payments and high level interconnection/ imports in the Italian case suggests that the two, rather than opposite, have complementary

43 Guido Bortoni, ‘Annual report on the state of services and regulatory activities’, presentation of 26 June 2013, Rome, available at http://www.autorita.energia.it/allegati/relaz_ann/13/annualreport_2013.pdf, accessed 1 February 2015, pp 3–4. 44 Communication from the Commission, Delivering the internal electricity market and making the most of public intervention, C(2013) 7243 final (November 2013 Communication), in particular, Commission staff working document, Generation Adequacy in the internal electricity market—guidance on public interventions, 5 November 2013 (Generation Adequacy SWD). 45 ACER’s Report (n 10) p 5. See chapter 2 for a discussion. 46 Communication from the Commission, Guidelines on State aid for environmental protection and energy 2014–2020 [2014] OJ C 200/1–55 (EEAG 2014–2020) para 234(a).

312

Italy

aspects, with capacity payments acting as an ‘insurance policy’ against failure of domestic as well as neighbouring countries’ generation capacity.

17.4.2.2 The external dimension: The relation with Switzerland The Generation Adequacy SWD stresses that ‘Member States’ generation adequacy assessments need to take account of existing and forecast interconnector capacity as well as the generation adequacy situation in neighbouring Member States.’47 This could be read as including neighbouring countries outside the EU. Indeed, in the EEAG 2014–2020, the Commission recalls that participation in RES promotion schemes ‘should in principle be open to other EEA countries and Contracting Parties of the Energy Community to limit the overall distortive effects.’48 The same rationale could apply to capacity mechanisms. The Italian experience provides other interesting insights in this respect. First of all, since the beginning of the liberalization of the electricity market, the Italian authorities have tried to cooperate closely with the Swiss authorities, for instance by a process of coordinated allocation of interconnection capacity. Second, in the Green Network case, brought before the CJEU as a reference for preliminary ruling, the Italian Consiglio di Stato, the highest administrative Court, has raised a number of questions pertaining to, in essence, the competence of the Union to conclude international agreements to recognize the origin of electricity from non-Member States as coming from a RES.49 47

Generation Adequacy SWD (n 44) p 6. EEAG 2014–2020 (n 46) para 122. 49 Case C-66/13 Green Network SpA v Autorità per l’energia elettrica e il gas (Green Network) [2013] OJ C 147/6. The questions referred read: ‘1. Is it inconsistent with the correct application of Articles 3(2) and 216 TFEU—according to which the Union has exclusive competence for the conclusion of an international agreement when that conclusion is provided for in a legislative act of the Union or is necessary to enable the Union to exercise its internal competence, or in so far as its conclusion may affect common rules or alter their scope, with the twofold consequence that, first, the power to conclude with non-Member States agreements that affect common rules or alter their operation, or [affect] a sector completely governed by Community law and for which the Union has exclusive competence, is centralised within the European Union itself and, secondly, that such authority no longer resides individually or collectively with the Member States—and of Article 5 of Directive 2001/77/EC, for a national provision ([Article] 20(3) of Legislative Decree No 387 of 2003) to make the recognition of the guarantees of origin issued by third States subject to the conclusion of an appropriate international agreement between the Italian State and the third State in question? 2. Are the national rules at issue inconsistent with the correct application of the abovementioned Community rules, when the Non-Member State is the Swiss Confederation, linked to the European Union by a free trade agreement concluded on 22 July 1972 and entered into force on 1 January 1973? 3. Is it inconsistent with the correct application of the Community rules referred to in question (i) for the provision of national law, contained in Article 4(6) of the Ministerial Decree of 11 November 1999, to lay down that, when electricity is imported from non-Member States of the European Union, acceptance of the application is conditional upon the conclusion of an agreement between the National Grid Manager and an equivalent local authority determining the detailed rules for the necessary checks? 4. In particular, is it inconsistent with the proper application of the Community rules at issue for the agreement under Article 4(6) of the Ministerial Decree of November 1999 to consist of a merely tacit agreement, never set out in official documents and the subject of a mere statement by the appellant, which is unable to provide details of its essential elements?.’ 48

17.5 Conclusion

313

In November 2014 the Court handed it down its judgment, holding that the European Union does enjoy exclusive external competence.50 The Consiglio di Stato is due to rule on the matter in the second half of 2015. Given that the ability to rely on the RES origin of the electricity injected into Italy may have an impact on imports,51 this preliminary ruling is bound to affect the pattern of electricity trading.

17.4.2.3 Conclusion on the EU dimension The capacity mechanism in Italy is yet to be finalized. Thus, it is too early to make a full assessment as to its compatibility with the EEAG and with the free movement rules, let alone with the antitrust provisions. However, even at this preliminary stage, the Italian experience raises a number of interesting points as to the possibility of reconciling capacity mechanisms with the internal energy market. Italy also provides a useful example of how to successfully integrate non-EU Member States into the electricity market.

17.5 Conclusion In 2003 Italy adopted a capacity mechanism (essentially consisting of a payment to eligible generators) to respond to a capacity gap situation, compounded by a high reliance on low-cost imports. In 2013, more than 30 per cent of the electricity produced in Italy came from RES. As a consequence, Italy started a transition towards a capacity mechanism which also caters for managing the intermittency associated with RES. In 2014 Italy is in the midst of this transition, and new rules for the existing and final capacity mechanism are still to be defined. Nevertheless, with a decade of experience in capacity mechanism, Italy provides useful insights on the interplay between capacity mechanisms and interconnection/imports and coordinating with neighbouring countries.

50

Judgment of the Court of 26 November 2014 in Green Network (n 49). Italian law (Legislative Decree No 79 of 1999) require importers of non-RES electricity to buy green certificates in proportion to the imports; importers that can show the RES origin of imports are exempted. Thus, if an importer cannot rely on the RES origin, it will have to bear the additional cost of purchasing green certificates. 51

18 Netherlands Marinus Winters

18.1 Introduction The Netherlands has an energy-only market. Several electricity generators have indicated that generators of conventional power should be paid for the capacity made available, since gas-fired power plants are loss-making at the moment as a result of the high gas prices and the low electricity prices. The Dutch government does not have any intention of implementing a system of capacity mechanism, since the generation adequacy in the Netherlands is not at stake. This chapter is structured as follows. Section 18.2 focuses on the generation adequacy situation in the Netherlands. It also describes the regulatory framework and reports key market figures, essential for the security of supply assessment. Section 18.3 argues that the Dutch energy-only market model is fully in line with the EU approach to capacity mechanism. Section 18.4 concludes.

18.2 Setting the scene 18.2.1 Market characteristics The generation portfolio in the Netherlands can be characterized as a diversified portfolio, provided that more than half of the production capacity consists of gasfired capacity. The Netherlands has sufficient capacity to cover the Dutch peak demand and therefore there is an overcapacity on the market. This overcapacity will grow in the coming years, since it is expected that the installed capacity of conventional power as well as RES power will increase until 2020.

18.2.2 Regulatory framework The production, distribution, trade, and supply of electricity is governed by the Dutch Electricity Act 1998 (Electricity Act).1 The Electricity Act implements the 2009 Electricity Directive2 and mainly contains obligations for grid operators. The Dutch competition authority (Autoriteit Consument & Markt, ACM) enforces the obligations in the Electricity Act. 1

Elektriciteitswet, 2 July 1998, Stb. 1998, 427, as amended Stb. 2013, 325 (Electricity Act). Directive 2009/72/EC of the European Parliament and of the Council of 13 July 2009 concerning common rules for the internal market in electricity and repealing Directive 2003/54/EC [2009] OJ L 211/55 (2009 Electricity Directive). 2

18.2 Setting the scene

315

The market for the wholesale and production of electricity in the Netherlands is fully liberalized since the late nineties. The Electricity Act does not contain regulations in relation to capacity mechanisms. Therefore the wholesale and production market is an energy-only market, where the generators are not paid for the capacity that is available to the market. In relation to the balancing market it should be noted that there used to be an obligation for generators to keep at least 1 per cent of their production capacity available for the primary reserve. The generators did not receive any compensation for the availability of this production capacity. As of January 2014 this obligation is no longer applicable. This means that the Dutch TSO (TenneT), will auction the primary capacity on the market. This shift from a mandatory to a free-market model is in line with the 2009 Electricity Directive, which stipulates that the TSO shall procure the energy used to cover energy losses and reserve capacity according to transparent, nondiscriminatory and market-based procedures.3 TenneT has decided to operate these reserve capacity auctions on a German platform of several TSOs.4 The winning bidders on this auction will only get a capacity payment and not a payment for the energy. Pursuant to the Electricity Act the Minister of Economic Affairs has an obligation in relation to the monitoring of the security of supply. This obligation includes the collection, analysis, and process of data in relation to the security of supply. In this respect the Minister of Economic Affairs has the authority to order that generators (and/or suppliers and traders) provide information to carry out this task. This monitoring obligation is an implementation of Article 4 of the 2009 European Electricity Directive5 and covers the balance of supply and demand on the national market, the level of expected future demand, and the envisaged additional capacity being planned or under construction. The Minister of Economic Affairs has the authority to issue an order to TenneT to carry out activities in relation to the minister’s task. According to the legal history the Minister of Economic Affairs can ask the TSO to carry out these activities, since the TSO has more technical knowledge to analyse this data. On the basis of the aforementioned monitoring obligation the Minister of Economic Affairs is obliged to publish a Monitoring Report on the Security of Electricity and Gas Supply (Monitoring Report) each year. In October 2014 the Minister of Economic Affairs published the last Monitoring Report in 2014.6 Apart from an analysis of the security of supply, this report also contains an analysis of the provision of electricity. This means that the Minister of Economic Affairs has assessed the required network capacity and the technical condition of the grid in terms of the interruptions.

3

2009 Electricity Directive (n 2) Art 15(6). See http://www.regelleistung.net, accessed 1 February 2015. 5 2009 Electricity Directive (n 2). 6 Ministry of Economic Affairs, Monitoring Report on the Security of Electricity and Gas Supply (Monitoringsrapportage Leverings- en Voorzieningszekerheid Elektriciteit en Gas 2014), 24 October 2014, available at http://www.rijksoverheid.nl/documenten-en-publicaties/rapporten/2014/10/23/monitoringsrapportageleverings-en-voorzieningszekerheid-elektriciteit-en-gas-2014.html, accessed 1 February 2015 (Monitoring Report 2014). 4

316

Netherlands

18.2.3 Generation adequacy As indicated earlier the market for the wholesale and production of electricity has been liberalized since the late nineties. The largest generators, which were previously owned by Dutch Provinces and Municipalities, have been acquired by foreign companies in the past years and these generators are currently owned by GDF/Suez, RWE/Essent, Vattenfall/Nuon, and E.ON. On the basis of public information it can be concluded that in 2014 the total installed capacity exceeds 26 GW and this generation portfolio has the following characteristics, illustrated in Table 18.1 below. In addition to the production capacity indicated in Table 18.1 there is a lot of decentralized gas-fired power capacity in the form of CHPs, which may not be taken into account in the earlier overview. These CHPs are mostly installed at the heavy industry and the horticultural sector. Please note that, according to the Monitoring Report 2014, total operational capacity in the Netherlands amounts to 28.7 GW, which consists of 24.2 GW thermal power and 3.4 GW of renewable power.7 The installed generation capacity in the Netherlands has grown in the past years and it will further grow in the coming years. Big projects, which will come online are the 1,600 MW coal-fired power plant of RWE in the Eemshaven-area, an onshore and near shore wind farm in the Noordoostpolder of 450 MW, which will be developed by several developers (among others, RWE) and the 600 MW offshore wind farm, which will be developed by Northland Power, Siemens, Van Oord, and HVC. The development of mainly wind energy will be stimulated in the coming years. The Dutch government has given concrete commitments to facilitate the development of wind energy. In relation to onshore wind the Dutch Provinces have agreed that Table 18.1 Installed capacity per generation source, 2014 Power plant type

Installed capacity (MW) in 2014

Gas Coal Nuclear Other Wind onshore Wind offshore PV

16,057 6,098 492 791 2,4348 180 7629

Total

26,397

Source: Author’s own table based on data notified to TenneT and published on http://www.tennet.org, accessed 1 February 2015, unless stated otherwise.

7

Monitoring Report 2014 (n 6) p 6. Compendium, ‘Wind power in the Netherlands 1990-2013’, available at http://www.com pendiumvoordeleefomgeving.nl/indicatoren/nl0386-Windvermogen-in-Nederland.html?i=6-38, accessed 1 February 2015. 9 Letter of Minister of Economic Affairs to Parliament, 15 April 2015, ref. DGETM-ED / 15047549. 8

18.2 Setting the scene

317

ultimately in 2020 areas will be allocated, where 6,000 MW of wind energy can be developed. In relation to offshore wind it should be noted that the government has entered into the ‘National Energy Agreement’ with many interest groups in the energy sector. According to the National Energy Agreement the Minister of Economic Affairs will tender the allocation of subsidies for the development of wind energy in several tender rounds, which equals to an installed capacity of 4,500 MW of offshore wind in 2023. On the other hand, it has been agreed in the National Energy Agreement that five older coal-fired power plants will be decommissioned ultimately in 2017. In this respect it should be noted that the generators have entered into the National Energy Agreement under the condition that ACM confirms that the agreement to decommission the older coal-fired power plants is not in breach of Article 101 TFEU. However, ACM is of the opinion that this agreement may be in breach of Article 101 TFEU, since the National Energy Agreement is in fact a joint agreement to limit the production capacity.10 This limitation of the production capacity may lead to an increase of the market price on the wholesale market of 0.9 per cent over the period 2016–2020. ACM is furthermore of the opinion that the environmental advantages of the decommissioning of the older coal-fired power plants do not outweigh the disadvantages of the expected price increase. The parties to the National Energy Agreement are therefore currently discussing alternative agreements to reach the goals of the National Energy Agreement, but it is still expected that approx 3,400 MW of coal-fired capacity will be decommissioned by 2017. The National Energy Agreement shows that government intervention appears to contradict other interests such as competition law. Even if the older coal-fired power plants are decommissioned it is expected that the installed capacity in the Netherlands is sufficient to cover the Dutch peak demand of around 21,000–22,000 MW.11 According to the Monitoring Report 2014, the Minister of Economic Affairs forecasts that, in 2018, there will still be a (firm) generation surplus of 3.2 GW, despite the planned closure of coal-fired generation.12 This expectation is supported by ENTSO-E’s Report.13 According to ENTSO-E, the Netherlands has a remaining capacity14 of between 9.5 GW and 14.5 GW in 2020 (depending on the different scenarios and seasons). Finally it should be noted that the Dutch electricity market is coupled with the Central-West European region (Belgium, France, Germany, and Luxembourg), Norway, and the UK. ACM is also amending the market regulations in order to create the North-western European market. The creation of a wider market will also facilitate the 10

ACM, ACM analysis of closing down 5 coal power plants as part of SER Energieakkoord, Consultation document, 26 September 2013, available at https://www.acm.nl/en/publications/publication/12082/ACManalysis-of-closing-down-5-coal-power-plants-as-part-of-SER-Energieakkoord/, accessed 1 February 2015. 11 TenneT, Quality and Capacity Plan 2010–2016, available at http://www.tennet.eu/nl/about-tennet/ news-press-publications/publications/technical-publications/quality-and-capacity-plan-2010-2016.html, accessed 1 February 2015. 12 Monitoring Report 2014 (n 6) p 6. 13 ENTSO-E, Scenario Outlook and Adequacy Forecast 2014–2030, June 2014 (ENTSO-E’s Report). 14 ENTSO-E defines the remaining capacity as the difference between the reliably available capacity and load at reference point. The reliably available capacity is the net generating capacity minus the unavailable capacity.

318

Netherlands

security of supply and may also provide chances to generators in the Netherlands to sell the electricity in the neighbouring countries. It can be concluded that the security of supply is not at stake in terms of the installed capacity. However, as indicated earlier a substantial part of the Dutch production portfolio consists of gas-fired power plants. The generators with these gas-fired power plants currently face difficult times, since the gas prices are high and the electricity prices are low as a result of abundant renewable electricity, which is generated in Germany and sold on the Dutch market. Therefore generators with gasfired power plants are considering the mothballing or decommissioning of gas-fired power plants. RWE is even considering connecting one of the larger gas-fired power plants, which is situated near the border, with the Belgian electricity grid. As a result of these market circumstances market players are therefore of the opinion that the wholesale and production market in the Netherlands should switch to another business model, in which generators are paid by the capacity made available to the market.

18.3 Energy-only market and the European dimension 18.3.1 Position of the Dutch government on capacity mechanisms As indicated in the introduction to this chapter, the Netherlands has an energy-only market, which is fully liberalized. The security of supply in the Netherlands is assured. Therefore the Dutch government does not have any plans to implement a capacity mechanism. The Dutch government is not in favour of introducing a capacity mechanism in the Netherlands or elsewhere in Europe. In reaction to the Commission’s November 2013 Communication,15 the Dutch government is of the opinion that capacity mechanisms are a ‘second-best’ option. The Dutch government furthermore indicated that before a capacity mechanism will be implemented, it is necessary to assess the security of supply situation in the relevant Member State and it should be assessed, whether alternative market-based mechanisms to secure the security of supply can be implemented, such as DSR, production capacity in other Member States or the extension of interconnection capacity. It should also be assessed, whether the market is sufficiently liberalized. According to the Dutch government, a well-functioning energy market will give the right price signals, which will assure the security of supply. If Member States nevertheless decide to implement a system of capacity payments, the Netherlands support the advice of the Commission, which means that Member States should implement a cost-effective system, which is technology neutral, open for producers from other countries, and temporary.16

15

Communication from the Commission, Delivering the internal electricity market and making the most of public intervention, C(2013) 7243 final (November 2013 Communication). 16 Second Chamber of Parliament, Parliamentary Papers, session 2013–2014, 22 112, No 1754, p 7.

18.3 Energy-only market

319

18.3.2 ACM’s position on capacity mechanisms In July 2013 ACM published a ‘vision document’ with the strategic priorities for the Electricity and Gas wholesale markets.17 In this vision document, ACM explained its view on capacity mechanisms. According to ACM, a system of capacity payments is not necessary in the Netherlands. ACM is also of the opinion that an active role of both ACM and the Ministry of Economic Affairs within Europe is very important, since the Netherlands cannot isolate itself from the negative effects of capacity mechanisms in surrounding countries. Further, ACM notes that the security of supply is an international issue, and is a strong supporter of a joint approach to the problem of security of supply. In this respect ACM is in favour of a harmonization of market incentives for renewable electricity, in which the production of RES based on proven technology across the EU should be part of the energy markets (including program responsibility). ACM further believes that strengthening the European market integration contributes to the security of supply. The efficient use of the existing flexible generation capacity will be promoted across borders and incentives for new investments will be increased. It is also important to increase the demand side response, initially focused on industrial customers directly engaged on the wholesale energy market. According to ACM the introduction of capacity mechanisms should only be considered if it is proven that the inclusion of a renewable energy market and the completion of the European market are not sufficient to guarantee the security of supply. At the same time ACM realizes that surrounding countries are introducing capacity mechanisms. ACM is however of the opinion that any discussion on capacity mechanisms should be seen in the light of the transition to a low-carbon marketmodel. A shared European vision should be developed on a roadmap towards a situation with 100 per cent (unsubsidized) renewable energy with a large share of wind energy and solar power, efficient markets, and an adequate level of security of supply. This market model is likely to differ from the current market model, since the marginal costs of the production of solar and wind power are practically zero, and there is an increasing need for flexibility. A timely vision of this future market model is moreover important to provide clarity to investors. In view of the foregoing, ACM, in the vision document, indicates that national capacity mechanisms with a structurally negative effect on the functioning of the internal energy should not be implemented rapidly and without consultation of surrounding countries. Although a harmonized European solution is ultimately preferable, national measures with a short-term nature aimed at the security of supply could be considered, provided that these measures are compatible with the European rules for the internal energy market. This means that a capacity mechanism should be accessible for generators from other Member States. 17

ACM, Visiedocument strategische prioriteiten E&G groothandelsmarkten, No 104195/27.O381, 23 July 2013, available at https://www.acm.nl/nl/publicaties/publicatie/11701/ACM-publiceert-haar-prioriteitenvoor-de-groothandelsmarkten-voor-elektriciteit-en-gas/, accessed 1 February 2015.

320

Netherlands

In a European context ACM will use its best efforts to come to a shared European vision in relation to the future market model and where necessary, any appropriate transitional measures. Finally ACM is of the opinion that the introduction of a capacity mechanism is a policy decision and is not a decision for the regulator.

18.4 Conclusion The Netherlands has an energy-only market, and the Dutch government does not have any intention of implementing a system of capacity mechanism, since the generation adequacy in the Netherlands is not at stake. Both the government and the energy regulator, ACM, consider that a capacity mechanism as a ‘second-best’ option, and security of supply should be first addressed by resorting to alternative solutions, which is in line with the EU approach.

19 Norway Jens Naas-Bibow and Catherine Ramstad Wenger

19.1 Introduction Norway has an energy-only market and does not intend to develop capacity mechanisms. Instead, Norway is relying on market mechanisms and interconnectors with neighbouring States in order to ensure security of supply. The Norwegian government is of the opinion that the current energy market should be strengthened and developed Europe-wide in order to minimize the risk of unstable energy deliveries arising from the increasing ratio of renewable energy production in Europe. The aim of this chapter is to explain the political and factual backdrop which makes Norway one of the few states arguing for a limited role for the capacity mechanisms. First, the market characteristics, regulatory framework, and generation adequacy is explained in section 19.2. Secondly, an overview of the Norwegian energy-only market is given in section 19.3. This section focuses on the key features of the common Nordic electricity market and the Norwegian-Swedish electricity certificates market. In addition, it sets out the enforcement of the national energy regulation. Finally, section 19.4 discusses the European dimension, hereunder the positions of the other Nordic States in relation to capacity mechanisms and energy-only market. Section 19.5 concludes.

19.2 Setting the scene 19.2.1 Market characteristics The electricity market in Norway is an energy-only market, and it is not on the political agenda to introduce capacity mechanisms. Rather, it is feared that the establishment of capacity mechanisms, even beyond Norway’s borders, could distort the market mechanisms Norway relies upon. The former Norwegian red-green coalition government (2005–2013)1 was strongly opposed to the development of capacity mechanisms in Europe as Norway is part of a well-established energy market which is relied upon in times where demand for energy exceeds production. The government’s concern was that capacity mechanisms would distort the energy-only market and impede investment in new interconnectors with neighbouring states. Despite not formally stating its position, the current right-wing government (2013–current)2 has made it clear that a well-functioning energy market 1 The Red-Green Coalition of the Labour Party, Socialist Left Party, and Centre Party governed Norway as a majority government from the 2005 general election until 2013. The Centre Party was the ‘green’ element of the alliance, and is a centrist agrarian party. 2 The right-wing government consists of the Conservative Party and the Progress Party.

322

Norway

with prices reflecting the actual market situation is of great importance3 and is currently working to ensure that interconnectors will be taken into account in possible capacity mechanisms. About 90 per cent of Norway’s electricity production is publicly owned by the government, county municipalities and local municipalities.4 Most of the electricity grid and cables are publicly owned5 and the Norwegian energy sector is closely regulated by the government. Despite this strong, public ownership, the sector has a robust legal framework which ensures the participation of the private sector and strengthens private market mechanisms. Norway’s electricity generation totalled 127.6 TWh in 2011 and 146.4 TWh in 2012, and the production stems mainly from renewable energy sources; 95 per cent from hydropower and the remaining 5 per cent from thermal and wind.6 The hydropower production is vulnerable due to variations in rainfall which lead to unpredictable levels of catchment in the reservoirs. In addition, factors such as temperature,7 electricity prices and consumer behaviour will affect actual consumption and therefore lead to shortage of electricity at times. This unpredictability and vulnerability has to a great extent been solved by relying on cross-border interconnectors. Norway’s neighbouring states produce energy from a variety of different sources, such as renewable sources (ie wind and hydro) as well as coal, gas, oil and uranium. These energy sources will normally be available when shortages in Norway’s reservoirs result in insufficient domestic electricity production. Thus, Norway depends upon its interconnectors to import electricity in order to ensure security of supply. However, interconnectors usually allow for both the import and export of electricity and in times of excess electricity production, Norway exports electricity.

19.2.2 Regulatory framework The regulatory framework governing the energy sector in Norway can be found in an array of different laws and regulations, and some of these will be explained in this section. It is, however, worth noting that a large part of the laws and regulation governing the Norwegian electricity sector have been broadly drafted and therefore lack precision to a certain extent.8 As a consequence, the legal framework indirectly grants extensive powers to the Norwegian Water Resources and Energy Directorate, a 3 See, for example, the speech by the Norwegian Minister of Petroleum and Energy, Tord Lien at the Winter Conference, March 2014, available at http://www.regjeringen.no/nb/dep/oed/aktuelt/taler_artikler/ minister/taler-og-artikler-av-olje–og-energimini/Innlegg-pa-vinterkonferansen-til-Energi-Norge.html?regj_ oss=1&id=754089, accessed 1 February 2015. 4 Official Norwegian Report, Hjemfall (NOU 2004:26) para 10.4.3. 5 Official Norwegian Report, Energi- og kraftbalansen mot 2020 (NOU 1998:11) para 2.2.3. 6 In 2011, 121.6 TWh of electricity production came from hydro-power and 6 TWh from thermal and wind. Statistical data available at http://www.ssb.no/en/energi-og-industri/statistikker/elektrisitetaar/aar/ 2013-03-20#content, accessed 1 February 2015. 7 As Norway is situated in Northern Europe the temperature varies greatly between summer and winter, but also within each season as Norway has both ‘cold’ winters and ‘warm’ winters. 8 Jens Naas-Bibow, Gunnar Martinsen et al, Energiloven med kommentarer (Gyldendal Norsk Forlag, 2011) p 15.

19.2 Setting the scene

323

Norwegian government agency responsible for regulating the country’s water and energy resources (NVE, energy regulator).9 NVE, the Financial Supervisory Authority of Norway, and the Norwegian Competition Authority are responsible for the market surveillance and supervision of the energy market. The Norwegian TSO, Statnett SF, is responsible for the operation of Norway’s transmission system as well as the interconnectors to Sweden, Finland, Russia, Denmark, and the Netherlands. The rules related to the organization and functioning of the electricity sector in Norway are set out in the Energy Act10 and the Energy Regulation11 which implements the 2009 Electricity Directive.12 The aim of the Energy Act and the Energy Regulation is to establish an efficient, competitive, and open electricity market and to prevent discriminatory actions by network companies. The Energy Act and the Energy Regulation require a legal and functional separation between network activities, production, and sales of electricity for vertically integrated companies13 with more than 100,000 customers. However, a recent report initiated by the Ministry of Petroleum and Energy14 proposes to apply the legal and functional separation to all vertically integrated companies—including those with less than 100,000 customers.15 The economic regulation of network activities is set out in the Energy Regulation16 and the Regulation for Reporting and Revenue Cap.17 The purpose of these regulations is to ensure a socially rational energy sector by enabling an effective energy market and effective management, utilization, and development of the electricity network. Procurement in the network distribution sector is regulated by the Procurement Act18 which implements the Directive 2004/17/EC.19 In accordance with the Procurement Act, procurements that exceed predefined thresholds in value must be carried out in accordance with the procedures laid out in the Procurement Act. 9

NVE was established in 1921 and is a department of the Ministry of Petroleum and Energy. Lov om produksjon, omforming, overføring, omsetning, fordeling og bruk av energi m.m. (energiloven), Lov-2013-06-14-53, available at http://lovdata.no/dokument/NL/lov/1990-06-29-50, accessed 1 February 2015 (Energy Act). 11 Forskrift om produksjon, omforming, overføring, omsetning, fordeling og bruk av energi m.m. (energilovforskriften), FOR-1990-12-07-959, available at http://lovdata.no/dokument/SF/forskrift/199012-07-959, accessed 1 February 2015 (Energy Regulation). 12 Directive 2009/72/EC of the European Parliament and of the Council of 13 July 2009 concerning common rules for the internal market in electricity and repealing Directive 2003/54/EC [2009] OJ L 211/55 (2009 Electricity Directive). 13 That is, undertakings engaged in both electricity distribution and production/supply of energy. 14 The report was written by an expert group which was initially formed by the previous red-green coalition government and continued under the current right-wing coalition government. The report is currently being circulated for comments by the market participants and it is not yet clear whether the proposal will be implemented. 15 Expert group report, Et bedre organisert strømnett, 2013. See Press release at https://www.regjeringen. no/nb/aktuelt/ekspertgruppe-om-stromnettet-utnevnt/id727246/, accessed 1 February 2015. 16 Energy Regulation (n 11). 17 Forskrift om økonomisk og teknisk rapportering, inntektsramme for nettvirksomheten og tariffer, FOR1999-03-11-302, available at http://lovdata.no/dokument/SF/forskrift/1999-03-11-302, accessed 1 February 2015 (Regulation for Reporting and Revenue Cap). 18 Lov om offentlige anskaffelser (anskaffelsesloven), LOV-1999-07-16-69, available at http://lovdata.no/ dokument/NL/lov/1999-07-16-69, accessed 1 February 2015 (Procurement Act). 19 Directive 2004/17/EC of the European Parliament and of the Council of 31 March 2004 coordinating the procurement procedures of entities operating in the water, energy, transport, and postal services sector [2004] OJ L 134/1. 10

324

Norway

The Settlement Regulation,20 together with the Energy Act21 and the Energy Regulation,22 imposes restrictions on a network company’s handling of information and is applicable to all DSOs with over 100,000 customers. The general rule states that DSOs must handle information in a way which does not give some energy producers a competitive advantage. In order to ensure the proper functioning of the energy sector, the Energy Act and the Energy Regulation set out the requirement for market participants to have the necessary licences in place. The different types of licences are: (a) sales licence, (b) facility licence and (c) area licence. NVE is the licensing supervisory body. In addition, Norway has several laws regulating the use of water which have an impact on the ways hydro production can be utilized. The Conservation Plans,23 the Common Plan for Water Systems24, the Industrial Licence Act,25 the Water Regulation Act,26 and the Water Resources Act27 are all part of the legal framework regulating hydropower production. The Water Resources Act implements the EU Water Framework Directive.28 Ownership of the water resources are regulated through (a) the rules related to licensing, (b) the doctrine of reversion, according to which the ownership to the waterfall reverts back to the State after a certain period, as well as (c) the right of pre-emption, which gives the existing shareholders in a water resource a right to buy shares offered for sale before they are offered to the public. According to the Industrial Licence Act29 waterfalls of a certain size must be two-thirds publicly owned. In general, therefore, hydropower producing entities are owned by municipalities. However, due to the fact that some private entities acquired waterfalls prior to the licence requirement came into place in the early 1900s, various ownership structures can be found in practice. Thus, some private entities have full ownership of waterfalls without licences.

19.2.3 Generation adequacy Since 1974, Norway has been a net exporter of electricity most of the time (three out of four years on average).30 In 2012 Norway’s electricity production totalled 146.4 TWh, 20

Forskrift om måling, avregning og samordnet opptreden ved kraftomsetning og fakturering av nettjenester, FOR-1999-03-11-301, available at http://lovdata.no/dokument/SF/forskrift/1999-03-11-301, accessed 1 February 2015 (Settlement Regulation). 21 22 Energy Act (n 10). Energy Regulation (n 11). 23 Verneplan I-IV, St.prp. nr. 4 (1972–73); St.prp. nr. 77 (1979–80); St.prp. nr. 89 (1984–85); St.prp. nr. 118 (1991–92). 24 Samlet plan for vassdrag, St. meld. nr. 60 (1991–92). 25 Lov om erverv av vannfall mv. (industrikonsesjonsloven), LOV-1917-12-14-16, available at http:// lovdata.no/dokument/NL/lov/1917-12-14-16, accessed 1 February 2015 (Industrial Licence Act). 26 Lov om vasdragsreguleringer (vassdragsreguleringsloven), LOV-1917-12-14-17, available at http:// lovdata.no/dokument/NL/lov/1917-12-14-17, accessed 1 February 2015 (Water Regulation Act). 27 Lov om vassdrag og grunnvann (vannressursloven), LOV-2000-11-24-82, available at http://lovdata. no/dokument/NL/lov/2000-11-24-82, accessed 1 February 2015 (Water Resources Act). 28 Directive 2000/60/EC of the European Parliament and of the Council of 23 October 2000 establishing a framework for the Community action in the field of water policy [2000] OJ L 327/1 (EU Water Framework Directive). 29 Industrial Licence Act (n 25). 30 Statistical data, available at http://www.statnett.no/Drift-og-marked/Data-fra-kraftsystemet/Nokkeltall1974-2012/, accessed 1 February 2015.

19.3 Energy-only market

325

the electricity export was 22 TWh and the import was 4.4 TWh, ensuring 17.6 TWh of net export of electricity.31 In 2011 Norway’s electricity production totalled 125 TWh, exports came to 14.3 TWh while imports totalled 11.3 TWh, making the net export of energy 3 TWh.32 However, in 2010, despite total energy production of 123 TWh, Norway was net electricity importer of 7.6 TWh.33 Over the years production and consumption have varied greatly, due to several factors. Energy consumption is affected by electricity price, the standard of dwelling, number of residents in each dwelling, and variations in temperature, especially in the wintertime. Variations in energy production result from variations in catchment; some years are dry and others wet. Despite variations in rainfall and subsequent differing levels of catchment, which at times lead to inadequate capacity, the Norwegian government is of the opinion that a strong energy market with interconnectors to neighbouring states is the best solution to ensure generation adequacy. The available import capacity is of great importance in ensuring security of energy supply.34 Thus, Norway’s security of supply has been ensured despite times of inadequate generation as it imports electricity from neighbouring states. Workable interconnectors are therefore key in ensuring a secure and stable supply of energy for Norway.

19.3 Energy-only market 19.3.1 Background Norway was the first Nordic State to deregulate its electricity market, with the adoption of the Energy Act in 1990.35 The Energy Act came into force on 1 January 1991 and required that separated accounts must be kept for the network activities and generation and energy sales activities in order to prevent cross-subsidization (accounting unbundling). It also required network owners to offer equal, non-discriminatory tariffs to electricity suppliers and end users. As a result of the deregulation, Statnett Marked, a free market and liquid market place for the power industry was established in December 1992.36 Prior to this, the Norwegian power system was organized by the Norwegian Power Pool, a membership organization that operated as a service for Norwegian power producers.37 The establishment of Statnett Marked initiated the beginning of the Nordic power exchange adventure,38 and the Nordic electricity market was formally opened up to regional competition with the establishment of Nord Pool ASA in 1996.39 Market liberalization 31

32 33 Statistical data (n 30). Statistical data (n 30). Statistical data (n 30). NVE, Driften av kraftsystemet 2012, Rapport No 44 (2013) p 7. 35 Finland liberalized its energy market in 1995. Sweden liberalized its energy market in 1996. Denmark started the liberalization of its energy market in 1999 and opened its electricity market for all consumers in 2003. 36 Hans-Arild Bredesen and Terje Nilsen, ‘Power to the People—the first 20 years of Nordic powermarket integration’ (Nord Pool Spot & NASDAQ OMX Commodities, 2013) p 30. 37 The Norwegian Power Pool was established 1 January 1971 and was to some extent continued under Statnett Marked. 38 Bredesen and Nilsen (n 36). 39 For more information on Nord Pool, see section 19.3.2.2 below. 34

326

Norway

ensured less divergence in the price of electricity between different regions in Norway. It also promoted investment in new production capacity where it was most needed.40 The liberalization of the electricity market has in many ways taken place in response to EU energy legislation. However, the liberalization occurred in conjunction with a general shift in Norway for the de-centralization and liberalization of the Norwegian economy in general. Today, the position of the Norwegian government in energyrelated matters is still aligned to a great extent with that of the Commission.

19.3.2 Key features 19.3.2.1 Cross-border trade of electricity Non-renewable energy production is vulnerable to peak demand as it is expensive and time consuming to regulate and build the non-renewable energy plants. Hydro power, due to its capacity to be stored, is efficient and cost effective in times of peak demand over short periods of time. The production of hydro power is, however, limited by the water reserve and is therefore vulnerable to variations in catchment. The Norwegian government is of the opinion that the shortcomings of hydro electricity production are to a great extent solved by the establishment of cross-border trade in electricity. Cross-border trade is seen as important by the Norwegian government as it reduces the inherent vulnerabilities of hydro power and ensures greater stability and security of electricity supply. Norway’s first interconnector was established in 1960 and connected Norway with Sweden. Since then, Norway has established interconnectors with Denmark, Finland, the Netherlands, and Russia. The total interconnection capacity for Norway today is around 5400 MW, and the government is currently negotiating with relevant authorities in Germany and the UK for a possible expansion of its transmission grid to these states.

19.3.2.2 The common Nordic electricity market The common Nordic electricity market was founded in 1996 when Norway and Sweden established Nord Pool. Finland entered Nord Pool in 1997 and Denmark entered the Nordic electricity market in 1999–2000.41 Estonia joined in 2010 and Lithuania in 2012. Nord Pool is an organized market for trade in electricity, and consists of several bidding areas.42 Bidding areas are geographical areas spanning an electrically unbroken part of the grid in the actual market area.43 In 2002, trading for physical electricity delivery was demerged into a separate daughter company, Nord Pool Spot AS. Trading 40 Before, the electricity price differences had led to investments in new production capacity in areas with high prices. The result was that expensive investments in areas with high price were preferred to less costly investments in the areas with cheaper price. 41 The western part of Denmark entered the market in 1999 and eastern part of Denmark entered the Nordic electricity market in 2000. 42 Bidding areas/bidding zones are also referred to as price areas/price zones. 43 For more information about bidding areas/price zones, see point (a) in this section.

19.3 Energy-only market

327

for physical electricity delivery can be in the form of wholesale or retail (directly to consumers). If trade is wholesale it is arranged through bilateral contracts (less than 30 per cent) or on the electricity market Nord Pool Spot (more than 70 per cent). The wholesale market is regulated by NVE. Trading in financial electricity continues to take place through Nord Pool, and is done for risk management and price hedging purposes, without the actual delivery of electricity. The form of the contracts varies between future- and forward contracts, options, and contracts for difference but does not exceed a period of six years. The trade takes place on the marketplace NASDAQ OMX Commodities.44 (a) Wholesale trade through Nord Pool Spot AS The Nordic electricity exchange Nord Pool Spot primarily serves the players on the wholesale market for electricity in the Nordic and Baltic States (Norway, Sweden, Denmark, Finland, Estonia, and Lithuania). Trade in Nord Pool Spot can be arranged a day ahead through Elspot or intra-day through Elbas. Nord Pool Spot is owned by the Nordic TSOs, and in Norway, NVE is responsible for the overall surveillance of this market. Nord Pool Spot sets the area prices as well as the daily reference price for the Nordic region (the system price) calculated according to supply and demand bids for electricity in the Nordic market45 to be delivered the following day. In order to set the system price, the transmission network constraints46 are disregarded. If there are no constraints, the electricity prices in all of the Nordic States (ie all bidding areas) are equal to the system price. However, if there are constraints in the network the Nordic electricity market subdivides into separate price areas. Norway is currently divided into five price areas, Sweden into four, Denmark into two areas and Finland, Estonia and Lithuania consist of one area each.47 The price does not necessarily have to differ between the different areas, but normally it does according to the local production situation. As explained earlier, Nord Pool Spot sets an area price which takes into account possible bottlenecks in the transmission network between the price areas/bidding zones. The area price in areas with excess electricity production will therefore be lower than the system price whereas the area price in areas with insufficient electricity production will be higher than the system price. However, the goal is that electricity should be able to flow freely between areas. Market mechanisms ensure that the electricity runs from the area with low price to the area with high price, thus evening out the price differences.

44 Information regarding NASDAQ OMQ is available at http://www.nasdaqomxnordic.com/, last accessed 1 February 2015. 45 Other factors, such as the price for electricity outside the Nordic market, are also decisive in setting the system price. 46 Transmission network constraints are also commonly known as transmission bottlenecks. 47 For more information on Nord Pool Spot, go to http://www.nordpoolspot.com/Market-data1/ Elspot/Area-Prices/ALL1/Hourly/, accessed 1 February 2015.

328

Norway

(b) Retail trade Almost 90 per cent of the retail market is household customers whereas the rest are commercial and industrial customers. The retail market is highly transparent and competitive as customers have full freedom to change their electricity supplier and information regarding products and prices are easily available. (c) The regulation power market In addition to financial and physical trading, trading can take place on the regulation power market, which is directed by Statnett. The regulation power market was established to ensure a balance between supply and demand of electricity in real time. The balance in the regulation market is achieved by Statnett SF, which ensures that producers either buy back electricity or will buy more electricity from the producers at a price below or above market price respectively. Only producers with the ability to respond quickly are active participants in this market.

19.3.2.3 The Norwegian-Swedish electricity certificates market Norway and Sweden entered into a treaty for establishing a common market for electricity certificates on 29 June 2011.48 The market has been operational since 1 January 2012 and its aim is to increase the production of renewable energy. In Norway the certificates market is regulated by NVE, whereas in Sweden it is the Swedish Energy Agency. The treaty has been implemented into Norwegian law with the enactment of the Electricity Certificate Act49 and the Electricity Certificate Regulation.50 The treaty shall cease to apply on 1 April 2036.51 The goal of the electricity certificates market is to finance the difference in market price and the cost of new production so as to promote investment in renewable energy. The common goal is to produce 26.4 TWh of renewable energy by 2020.52 Currently, investment has taken place mainly in Sweden. In 2013 4.1 TWh was invested in Sweden and 0.6 TWh in Norway.53 Electricity producers receive electricity certificates from Statnett SF, who is responsible for registering, issuing and deleting certificates. The certificates are given to the renewable energy producers according to their respective production,54 48

Agreement between the government of the Kingdom of Norway and the government of the Kingdom of Sweden on a common market for electricity certificates (Swedish-Norwegian Treaty). An unofficial translation is available at http://www.statnett.no/Global/Dokumenter/Kraftsystemet/Elsertifikater/swedish_norwe gian_treaty.pdf, accessed 1 February 2015. 49 Lov om elsertifikater, LOV-2013-12-13-126, available at http://lovdata.no/dokument/NL/lov/201106-24-39, accessed 1 February 2015 (Electricity Certificate Act). 50 Forskrift om elsertifikater, FOR-2011-12-16-1398, available at http://lovdata.no/dokument/SF/ forskrift/2011-12-16-1398?q=lov+om+elsertifikater, accessed 1 February 2015 (Electricity Certificate Regulation). 51 The Swedish-Norwegian Treaty (n 48) Art 15(1). 52 The Swedish-Norwegian Treaty (n 48) Art 2(1). 53 NVE, Elsertifikater: Statusrapport (July 2013) available at http://www.energinorge.no/getfile.php/ FILER/NYHETER/ENERGIPRODUKSJON/elsert_2kv13_02092013.pdf, accessed 1 February 2015. 54 1 MWh equals one electricity certificate.

19.3 Energy-only market

329

and the producers will later sell the certificates to electricity suppliers and certain electricity users. Power suppliers and certain electricity users55 are legally obliged to purchase a set amount of certificates, but may choose to purchase them in Norway or Sweden. NVE and its Swedish counterpart set quotas for how many certificates end-users need to purchase by kWh electricity consumed. The costs incurred for purchasing the electricity certificates by the power suppliers ripples down to each single consumer in the form of an additional cost to their electricity bill. Trading of electricity certificates across borders was believed to ensure a more effective utilization of renewable energy sources than if only national markets had been established. The certificates are given regardless of whether the power plant is located in Sweden or Norway and do not differ between what type of renewable energy source is used. Thus, investment in renewable energy generation was intended to take place where profitability was highest. The following plants are entitled to electricity certificates: (a) power plants based on renewable energy sources with construction start after 7 September 2009, (b) existing power plants expanding their production on a permanent basis (construction start after 7 September 2009), and (c) hydroelectric power stations with installed capacity up to 1 MW that had construction start after 1 January 2004. The price for the certificates is determined by supply and demand of electricity. Both yearly demand for electricity and the set certificate quota for that year are taken into account in determining the demand. The quota of electricity certificates to be bought by the energy suppliers is given independent of the electricity certificates issued. Market mechanisms are then relied upon in order to ensure that the most economically efficient investments are taking place when the electricity certificate price is high. In short, large-scale investments in new energy production are likely to result in a greater number of certificates and will most likely lead to lower price of each certificate. Conversely, few power plants under construction are likely to cause rising certificate prices until the prices reach a level where the energy market will attract new investors. The Norwegian government expressed its opinion in relation to the electricity certificate market in its non-paper of 16 May 2013; ‘the system is considered to be an effective use of one of the internal cooperation mechanisms under the renewable directive II’.56 However, market participants have questioned the effectiveness of the electricity certificate market and pointed at how a more relaxed tax regime in Sweden is promoting less energy efficient and costly investments to take place in Sweden rather than Norway.57 55 In general, the electricity users which have entered into bilateral supply agreements with the producers are obliged to purchase electricity certificates. 56 Norwegian views on European energy issues (non-paper, 16 May 2013) available at https://www. regjeringen.no/globalassets/upload/oed/foredrag20i20pdf20som20er20lagt20ut20pc3a520nettet/norway_nonpaper_20130522.pdf, accessed 1 February 2015. 57 Thema Consulting Group, ‘Sertifikatkraft og skatt –oppdatering’ (THEMA Report 2014–26, Executive Summary, May 2014) available at http://www.energinorge.no/getfile.php/FILER/NYHETER/EN ERGIPRODUKSJON/THEMA-rapport%202014-26%20Sertifikatkraft%20og%20skatt%20-%20oppdatering_ Sammendrag.pdf accessed 1 February 2015.

330

Norway

19.3.2.4 The electricity grid The electricity grids in Norway are divided into three levels: the central transmission grid, the regional grid, and the distribution grid. Under the Energy Act,58 entities conducting or acquiring transmission/distribution or retail business must obtain licences for both their grid installations and electricity sales. The Norwegian transmission grid is largely operated with modern technology, and market-based solutions59 have contributed to efficient price formation and a good utilization of the power system. The Nordic power system has been well integrated for several years. In ensuring a rational development of the system it has been important to ensure close cooperation with power grid companies, and this will continue to be of importance in years to come.

19.3.2.5 Enforcement of energy market regulations As Norway does not have a capacity mechanism, there are therefore no enforcement measures for capacity mechanisms in place. However, there are several governmental law enforcement agencies involved in the regulation of the energy markets: NVE, the Competition Authority, and the Financial Supervisory Authority of Norway. These bodies cooperate to ensure compliance with the laws and regulation of Norway’s energy-only market. NVE is responsible for the supervision, inspection, and reporting of the electricity market. NVE’s Energy and Regulation Department ensures supervision by issuing formal guidance or informal statements in relation to main areas important to securing an efficient electricity market. The areas of particular importance are network regulation and tariffs, quality of supply, metering and settlement, billing, supplier switching, neutrality and non-discrimination, and, finally, the obligations and powers of the TSO, Statnett SF. In order to monitor compliance NVE undertakes audits of the market participants and issues annual reports for these participants. These reports are publicly available on the NVE website.60 NVE also issues weekly, quarterly, and annual reports on developments in the Norwegian and Nordic electricity markets. Furthermore, NVE ensures compliance with the Electricity Certificate Act61 and with regulations establishing the Norwegian-Swedish electricity certificates market.62 The Electricity Certificate Act sets out different measures which can be taken in case of non-compliance. This includes revocation of the right to receive electricity certificates as well as financial and criminal penalties.63 Appeals against decisions will be dealt with by the Norwegian Ministry of Petroleum and Energy which is the final regulatory body 58

Energy Act (n 10). Examples of market based solutions are advanced metering systems (AMS) and the market place Nord Pool Spot. 60 All audit reports are available at http://www.nve.no/no/Sikkerhet-og-tilsyn1/Tilsyn-medkraftmarkedet/ accessed 1 February 2015. 61 Electricity Certificate Act (n 49). 62 For more detailed information regarding the electricity certificate market, see section 19.3.2.3. 63 Electricity Certificate Act (n 49) Arts 24–29. 59

19.4 European dimension

331

for this matter.64 In addition, the Electricity Certificate Act gives Sweden the right to receive information when necessary to ensure enforcement of the rights and obligations under the Act.65 Nord Pool Spot has, as a result of the obligations to its licence under the Energy Act, an obligation to establish appropriate procedures to monitor the behaviour of the market participants in the organized marketplace and Nord Pool’s Market Surveillance performs this task. Together with Nord Pool ASA’s Market Surveillance, they monitor the trading activities in the spot and derivatives markets at Nord Pool and conduct investigations of possible breaches of laws and regulations. The Competition Authority has the overall responsibility for effective competition between the actors in the energy market, and Norway has entered into a Nordic agreement for co-operation in competition cases.66 The treaty entered into force 1 April 2001. The Financial Supervisory Authority is responsible for the supervision of stock markets, authorized markets, clearing houses, and the securities registers for electricity contracts. Court procedures are also applicable in ensuring law enforcement in Norway’s energy sector.

19.4 European dimension 19.4.1 Acknowledging the EU context Despite not being part of the European Union, Norway is a part of the European Economic Area and most EU legislation will therefore also be implemented in Norwegian law. This is also the case with the EU energy law. Currently, Norway is in the process of amending parts of its energy law following the introduction of the Third Energy Package.67

19.4.2 Acknowledging the Nordic context As explained earlier, Norway is part of the common Nordic electricity market and has established an electricity certificates market with Sweden. Sweden, Denmark, and Finland hold similar positions to capacity mechanisms as Norway. The Nordic States believe that the electricity market functions well and that cross border trade contributes significantly to the improvement of security of supply. In addition, all Nordic States 64

65 Electricity Certificate Act (n 49) Art 23. Electricity Certificate Act (n 49) Art 23. Agreement between Denmark, Iceland, and Norway on cooperation in competition cases. Please note that Sweden joined at a later stage and is therefore not mentioned in the title of the treaty. 67 Directive 2009/72/EC of the European Parliament and of the Council of 13 July 2009 concerning common rules for the internal market in electricity and repealing Directive 2003/54/EC [2009] OJ L 211/55 (2009 Electricity Directive). Directive 2009/73/EC of the European Parliament and of the Council of 13 July 2009 concerning common rules for the internal market in natural gas and repealing Directive 2003/55/EC [2009] L 211/94 (2009 Gas Directive). Regulation (EC) 714/2009 of the European Parliament and of the Council of 13 July 2009 on conditions for access to the network for cross-border exchanges in electricity and repealing Regulation (EC) 1228/2003 [2009] OJ L 211/15 (2009 Cross-border Regulation). 66

332

Norway

believe that cross-border transfer capacity is the key to making the system more flexible and reliable.68 The Danish Energy Agency states in its response to the Commission’s 2012 Consultation Paper69 that ‘it is the Danish opinion that interconnectors shall be considered seriously important both in respect of a competitive electricity market and in respect of security of supply.’70 The Danish Energy Agency also points out that the ‘classic’ assessment of capacity balance does not seem to take sufficient account of the role of interconnectors. Denmark currently relies on importing electricity from its neighbouring states in times of peak demand when there is no wind power available in Denmark, as a result of the dismantling of the two big coal-driven power plants71 in January 2013. Sweden’s reply to the Consultation Paper highlights the risks associated with subsidizing fossil fuel and states that capacity mechanisms ‘may lead to reduction of trade between States, distortion of competition and that the cost of consumers increases’.72 Sweden argues that the fast development of renewable energy production is, and has been, important in improving security of supply and in reaching agreed climate targets. Sweden believes that capacity mechanisms should not be adopted as long as there are other cost efficient measures which do not distort price signals. Finland’s reply to the Consultation Paper also highlights the importance of considering the capacity mechanisms in light of the electricity market as there is ‘an apparent risk of deterioration of the energy-only target model of the electricity markets if uncoordinated national capacity remuneration schemes appear’.73 It is well known that Sweden and Finland both have strategic reserves. However, this fact does not undermine their position in relation to capacity mechanisms. It is generally believed that the phase-out of the strategic reserves is necessary, and that bidding zones, increased market integration and strengthened transmission networks should replace strategic reserves.

19.4.3 Assessment of capacity mechanisms under EU law Despite a change of government in October 2013, from the red-green coalition to a right-wing government,74 the current situation is not likely to change. The former Norwegian government was sceptical about the development of capacity mechanisms. Although the current government has not yet fully stated its political opinion on capacity mechanisms, the change of government will likely not impact on Norway’s position, as it is rather based on economic factors and hydropower production will continue to be of great importance in the foreseeable future. The right wing is also generally in favour of establishing markets and reducing state subsidies.

68 See the Nordic countries’ replies to the Commission’s 2012 consultation on generation adequacy, available at the Commission’s website at http://ec.europa.eu/energy/en/consultations/consultationgeneration-adequacy-capacity-mechanisms-and-internal-market-electricity, accessed 1 February 2015. 69 European Commission, Consultation Paper on generation adequacy, capacity mechanisms and the internal market in electricity (the 2012 Consultation Paper), available at the Commission’s website (n 68). 70 71 Danish Energy Agency’s reply (n 68) p 1. Enstedværket and Stigsnæsværket. 72 73 74 Sweden’s reply (n 68) pp 1–6. Finland’s reply (n 68) p 1. See n 1.

19.4 European dimension

333

In the former government’s reply to the 2012 Consultation Paper,75 the Norwegian Ministry of Petroleum and Energy described the views of the Norwegian government.76 In short, the former Norwegian government understood the other states’ concern of reducing risks related to security of supply. However, the government was of the opinion that ‘the market holds most solutions’ and emphasized the importance of allowing the market to work as efficiently as possible, whilst ensuring that any new regulatory interventions are well designed and effective. The Ministry believed that better solutions are found in the market than in capacity mechanisms. This was considered a better approach to rely upon, as well as developing efficient system operations and interconnections and working on demand side awareness. The Ministry highlighted the importance of removing market barriers, stating that technological, political, regulatory, and operational barriers should be removed. Finally, the Ministry argued that cross border trade of electricity creates added value through better utilization of energy systems, increased security of supply, and better integration of renewable energy.77 In addition to highlighting the benefits of a well-functioning energy market, the Ministry emphasized the risks of relying on non-market solutions such as capacity mechanisms. These risks include undermining the European Union goals for growth, decarbonization, and resource efficiency as well as any unnecessary interference in the energy market. The Ministry was concerned that non-market solutions such as capacity mechanisms could distort market behaviour and investment decisions in the internal market. The importance of relying on the internal market was emphasized and the Ministry stated that ‘if compelled to intervene’ in the market, ‘the measures should be as reversible, time limited and with as few possible unintended consequences as possible’. Thus, the Ministry stated that any market interventions to ensure security of supply should be limited and meet clear common criteria. In addition, the Ministry underlined the importance of careful assessment of the matter before concluding that the internal market will not ensure generation adequacy and security of supply. The Ministry fully agreed with the key message of the Commission’s 2012 Internal Energy Market Communication78 and stated that ensuring security of supply ‘is not just about avoiding blackouts, it is also about well-functioning electricity markets.’79 The development of capacity mechanisms is now a reality, and the Norwegian government is currently adjusting to this. Potential investment in interconnectors to the UK and Germany is, to a great extent, dependent upon how capacity mechanisms in these two countries are designed. From a Norwegian point of view, it is of great importance that electricity from Norway exported through new interconnectors would be able to participate in Germany’s and the UK’s capacity mechanisms. If this is not the case, introducing capacity mechanisms in these countries will undermine the economic viability of the planned interconnectors, and the investments in new interconnectors

75

76 77 2012 Consultation Paper (n 69). Norway’s reply (n 68). Norway’s reply (n 68) p 2. Communication from the Commission to the European Parliament, the Council, the European Economic and Social Committee and the Committee of the Regions: Making the internal energy market work, of 15 November, 2012, COM(2012) 663 final (2012 Internal Energy Market Communication). 79 2012 Internal Energy Market Communication (n 78) p 3. 78

334

Norway

are unlikely to go ahead. Thus, Norway has attempted to influence the British and German governments by entering into negotiations in relation to the interconnectors. These negotiations are currently ongoing. However, it is too early to know the outcome.

19.5 Conclusion The Norwegian energy-only market has developed over a period of many years and is continuously changing, reflecting new realities and situations. Despite heavy public ownership and governmental regulation, a robust legal framework ensures the functioning of market mechanisms and the involvement of the private sector. Norway has ensured security of supply by developing inter-governmental cooperation in a Nordic energy market and by relying on interconnectors with its neighbouring States. The Norwegian government is sceptical about the development of capacity mechanisms as it is believed it will influence negatively on the current energy-only market and possible future investments into interconnectors to other European States. Norway is an example of how cross-country cooperation and market mechanisms can ensure security of supply despite not relying on capacity mechanisms. In a world with an imminent need for a renewable future due to the global threat of climate change, the necessity of investigating whether the key to security of supply lies in the energy-only market cannot be underestimated. We hope that Norway’s experience with the energy-only market can contribute in inspiring other states and governments to follow this example.

20 Poland Małgorzata Sadowska

20.1 Introduction Poland’s electricity market remains energy-only, despite stopgap measures taken by the TSO in 2013–2014 to avert the risk of capacity shortages in the next three to four years. The possible introduction of a market-wide capacity mechanism has been discussed since the economic slowdown in 2009. Although the ongoing capacity debate has recently been heightened by market reforms in the UK and France, and similar proposals in Germany, the government has not yet produced any legislative proposal.1 Given that any capacity mechanism would lead to an increase in end customer prices, it is not expected that the government will take formal steps in that respect before the parliamentary elections in September 2015. This chapter surveys current developments in Poland with respect to generation adequacy and capacity mechanisms. After a brief overview of the market structure and its regulatory framework, section 20.2 explores the major adequacy challenge that Poland will face in the coming years. The newly employed capacity measures, as well as the potential adoption of a market-wide capacity mechanism, are discussed in more detail in section 20.3. Section 20.4 offers some thoughts about how to treat these developments from a broader European perspective. Section 20.5 concludes.

20.2 Setting the scene 20.2.1 Market characteristics The Polish electricity system,2 with 38.4 GW of installed capacity,3 is based on hard coal and lignite, the country’s main natural resources. The gross national electricity production amounts to 162.5 TWh, of which 87 per cent comes from coal,4 while the rest comes from renewable sources (5 per cent),5 natural gas (2 per cent) and other fuels.6 The gross national electricity consumption reaches 158 TWh.7 The average annual demand for power amounts to 21.8 GW, with the peak demand of 24.7 GW. 1

2 At the time of writing this chapter (January 2015). Based on 2013 market data. Energy Regulatory Office (URE), National Report, July 2014, p 33. 4 Hard coal—52%, lignite—35%. 5 Hydro-pump storage—2%. Wind, solar, biogas, and biomass (also from co-firing biomass with coal)—3%. 6 URE, National Report (n 3) p 12. 7 Since the political transformation in 1989, Poland has been a net exporter, and the surplus of export over import amounted to 4.5 TWh in 2013, mainly to the Czech Republic and Slovakia. The situation reversed in 2014, when more electricity was imported to the system than exported. This was due to lower electricity prices in Sweden and Germany, and the surplus of import over export amounted to 2.2 TWh. 3

336

Poland

Electricity is traded mostly through a power exchange (POLPX) and, to a lesser extent, through bilateral contracts. The role of POLPX has increased significantly since the introduction of the so-called ‘exchange obligation’ in 2010,8 and the total trading volume now accounts for 176.5 TWh.9 Following the government-led consolidation of the state-owned energy companies in 2006–2007, the market remains concentrated, with four big energy groups (PGE, TAURON, ENEA, and ENERGA). They all belong to the category of companies that the government treats as strategically important for the economy, and over which the state retains control. The state owns a controlling share in PGE (58 per cent), ENEA (51 per cent), and ENERGA (51 per cent), and remains the biggest shareholder in TAURON, with a 30 per cent share. PGE, ENEA, and TAURON are responsible for almost twothirds of electricity production and control more than half of installed capacity.10 The Polish electricity market is relatively isolated given its limited interconnection capacity. However, Poland takes part in two regional market initiatives. The first regional initiative is the North-Western European (NWE) market coupling, and even though Poland is currently only an observer in this project, day-ahead market coupling already exists on the SwePol Link, the DC interconnector with Sweden.11 Full membership of Poland in the NWE market based on Price Coupling of Regions is planned for 2016. The second initiative relates to market coupling in the Central and Eastern European (CEE) region, already existing between the Czech Republic, Slovakia and Hungary. Poland intends to join this market in the so-called Central East Europe FlowBased Market Coupling project (CEE FBMC). This will introduce day-ahead market coupling between Poland and the Czech Republic, Slovakia, Hungary, and Romania (‘M5 MC’).12

20.2.2 Regulatory framework The basic law governing the Polish energy sector (electricity, heating, and gas) is the 1997 Energy Law.13 According to this, security of electricity supply is the ability of the system to ensure both network security and generation adequacy.14 The energy policy is developed by the Ministry of Economy, which also coordinates its implementation.15 8 Since 2010, generators are obliged by law to sell no less than 15% of their electricity generated in a given year through the power exchange. Generators who receive compensation for stranded costs incurred after the early termination of long-term electricity supply contracts are obliged to sell all of their electricity through the power exchange. For more details, see section 20.3.2.3. 9 Trading volume of all four markets dedicated to electricity (Commodity Forward Instruments Market with Physical Delivery—CFIM, Day Ahead Market—DAM, Intra-Day Market, and Electricity Auctions). 10 Installed capacity and production, 2012: PGE—13 GW, 63.2 TWh; TAURON—5.5 GW, 22.1 TWh; ENEA—3 GW, 13 TWh; ENERGA—1 GW, 4.6 TWh. Shares in electricity fed into the grid, 2013: PGE— 39.3%; TAURON—13.6%; ENEA—8.1%; ENERGA—3.2%. 11 Day-ahead market coupling has already been introduced on the SwePol Link by means of bilateral cooperation between Poland and Sweden. There has been progress regarding the implementation of intraday market coupling and the allocation of long-term physical and financial transmission rights. 12 See Memorandum of Understanding w sprawie przyłączenia Polski i Rumunii do zintegrowanego mechanizmu market coupling na rynku dnia następnego, 11 July 2013. 13 Energy Law Act 1997, Journal of Laws of 2012, item 1059, as amended (Ustawa z dn. 10 kwietnia 1997 r., Prawo energetyczne, Dz.U. z 2012 r., poz. 1059 z późn. zm.). 14 15 Energy Law Act 1997 (n 13) Art 3. Energy Law Act 1997 (n 13) Art 12.

20.2 Setting the scene

337

Further, the Ministry exercises supervision over security of electricity supply.16 In particular, every two years, the Ministry issues a report on the results of monitoring the security of electricity supply covering five and fifteen-year periods (Ministry’s Report).17 The last Ministry’s Report comes from June 2013, and is discussed in more detail in section 20.2.3.1 below.18 Market monitoring in relation to the security of electricity supply is the task of the Energy Regulatory Office (URE, the energy regulator).19 In that respect, URE controls the activities of the TSO in terms of system security and reliability, and makes generation adequacy assessments on the basis of data provided by the TSO. URE is required to report on its monitoring to the Ministry every two years.20 In practice, URE includes its assessments in its National Reports, published on a yearly basis. The most recent National Report, which is referred to in this chapter, is from July 2014.21 In addition, URE conducts a long-term review of generation adequacy. This is based on the fifteenyear investment plans submitted to URE by generators every two years.22 URE’s most recent analysis of generation investments comes from November 2014 and covers the period of 2014–2028 (URE’s Investment Report, discussed in section 20.2.3.2).23 Polskie Sieci Elektroenergetyczne (PSE) S.A. is a 100 per cent state-owned and ownership unbundled TSO.24 As set out in the articles of the Energy Law, the TSO is responsible for the security of electricity supply and system security.25 In that respect, it is required to make generation adequacy forecasts, which are then submitted to the Ministry and URE, and contribute to the reports mentioned earlier.26

20.2.3 Generation adequacy Poland currently has a capacity surplus. According to the energy regulator and the TSO, the current levels of available capacity and operating reserve27 are adequate in terms of the operational security of supply.28 16

17 Energy Law Act 1997 (n 13) Art 12(2)(3). Energy Law Act 1997 (n 13) Art 15b(3). Ministry of Economy, Report on the Results of the monitoring of the security of electricity supply for period between 1.01.2011 and 31.12.2012, Warsaw 2013. 19 Energy Law Act 1997 (n 13) Art 23(2)(20)(f ). 20 Energy Law Act 1997 (n 13) Art 23(2a) and Art 23(2c). 21 22 See n 3. Energy Law Act 1997 (n 13) Article 16(20). 23 URE, Report on the investment plans in new generation capacity for years 2014–2028 (Informacja na temat planów inwestycyjnych w nowe moce wytwórcze w latach 2014–2028, Biuletyn URE 4(90), 25 November 2014). 24 In 2014, PSE was certified according to Article 11 of Directive 2009/72/EC of the European Parliament and of the Council of 13 July 2009 concerning common rules for the internal market in electricity and repealing Directive 2003/54/EC [2009] OJ L 211/55 (2009 Electricity Directive). 25 Energy Law Act 1997 (n 13) Art 9(2). 26 Energy Law Act 1997 (n 13) Art 9c(2)(16). 27 Operating reserve is excess capacity over and above electricity demand which is made available to the TSO, so that it can deal with actual demand being greater than forecast demand and/or plant unavailability. The required minimum level of the operating reserve is determined in the Grid Code (IRiESP) and amounts to 9% of projected peak demand. See Instruction of Operation and Exploitation of the Transmission Network (Instrukcja Ruchu i Eksploatacji Sieci Przesyłowej (IRiESP), version of 1 August 2014, Part 2 regarding conditions of the use and planned expansion of the network (Warunki korzystania, prowadzenia ruchu, eksploatacji i planowania rozwoju sieci), point 4.3.4.19(1)). 28 URE, National Report (n 3) pp 47–50. 18

338

Poland

That being said, there is an ongoing discussion, dating back to 2009, about introducing a capacity mechanism which would mitigate any potential risks of a future capacity shortage (henceforth referred to as the ‘capacity debate’). Most of the capacity concerns expressed in this debate so far relate to generation inadequacy, that is, the insufficient generation capacity to cover demand, rather than the lack of flexible resources to support intermittent RES. However, the flexibility problem is expected to emerge in the coming years, as the market becomes more advanced in developing renewable energy.29 At present, capacity concerns are caused by the current market conditions, particularly in the context of (1) forthcoming plant closures (ageing generation assets, stricter environmental standards) and (2) an unfavourable climate for new investments (low wholesale prices, the declining competitiveness of the coal business, regulatory uncertainty). First, a number of generation units will be decommissioned over the next few years. They amount to 5.2 GW of installed capacity and include mostly coal-fired power plants (more than 50 per cent), lignite-fired power plants (24.7 per cent) and installations co-firing coal and biomass (16.1 per cent). Most of these assets are decommissioned either due to old age (43 per cent) or non-compliance with stricter emission standards (43 per cent).30 With respect to ageing assets, it must be noted that approx. 45 per cent of generating installations are over thirty years old, and approx. 77 per cent are over twenty years old.31 Just to put this in context, the average lifetime of a coal-fired generation unit is forty to forty-five years, and a large share of the country’s generation was constructed during the 1970s. Stricter environmental standards, as imposed by the EU Directive on Industrial Emissions (IED),32 were transposed into national law in 2014.33 As a result, power plants which do not meet the IED standards have to be closed by 1 January 2016. A number of generation units will be temporarily exempted from the IED, but not beyond 2023. This includes the so-called ‘limited lifetime derogation’ under Article 33 of the IED, further discussed in section 20.3.1.1.34 Secondly, low wholesale electricity prices coupled with the rising costs of coalfired generation and regulatory uncertainty negatively affect investment in new

29

See section 20.2.3.2 below, last paragraph. URE’s Investment Report (n 23) pp 6–7. 31 Ministry of Economy, Energy Policy of Poland until 2050 (Draft, August 2014, Warsaw) p 13. 32 Directive 2010/75/EU of the European Parliament and of the Council of 24 November 2010 on industrial emissions (integrated pollution prevention and control) [2010] OJ L 334/17–119. 33 Act of 11 July 2014 amending the Environmental Law and Other Acts, Journal of Laws 2014, item 1101. This Act came into force on 5 September 2014, and was followed by the Regulation of the Minister of the Environment from 4 November 2014 regarding emission standards for certain types of installations, combustion plants, incineration and co-incineration plants, Journal of Laws 2014, item 1546. The Regulation has been in force since 22 November 2014. 34 There are three types of derogations from the IED Directive: (1) the so-called ‘limited lifetime derogation’ until 31 December 2023 for installations which committed to run less than 17,500 hours in the derogation period (78 units have applied for this derogation, which amounts to 13 GW), (2) derogation until 31 December 2022 for middle-sized heating plants (between 50 and 200 MW), which are used for district heating (70 units are eligible for this derogation, which amounts to 8 GW), (3) derogation until 30 June 2020 for installations listed in the so-called Interim National Plan, adopted by the government on 23 April 2014 and approved by the European Commission (67 GW, approx 73 units). 30

20.2 Setting the scene

339

generation, which is necessary to fill the capacity gap created by plant closures. Electricity prices dropped in 2009, reflecting the economic downturn. The trend continues,35 and in 2013 the average annual price of electricity sold on the competitive market was approx. 42 €/MWh (181.55 PLN/MWh).36 At this price level, keeping generation units in operation is unprofitable. Average electricity production costs are higher, ranging from 48 €/MWh (206 PLN/MWh) for a lignite-fired unit up to 82 €/MWh (351 PLN/MWh) for a coal-fired unit.37 As discussed later in this section,38 the most recent large-scale investment projects in coal generation have been motivated by government decision-making, and not by economic rationale. Frequent legislative changes have created regulatory uncertainty. Since its adoption in 1997, the Energy Law has been amended more than fifty times, resulting in three enactments of the consolidated text.39 Furthermore, in 2011 the Ministry of Economy initiated an ambitious reform of the Energy Law, which still has not been realized.40 Clashing interests and pressure from various lobby groups keep delaying the legislative process, which may add to regulatory uncertainty and investment risks.41

35 The average annual price of electricity sales on the competitive market in 2009–2013 in PLN/ MWh: 197.21 (2009); 195.32 (2010); 198.90 (2011), 201.36 (2012); 181.55 (2013). See the URE, National Report (n 3) p 36, table 10. Low wholesale prices are caused by overcapacity, lower electricity demand, cheap carbon prices, as well as the high share of depreciated units and relatively cheap lignite-fired units in the system. Increasing renewable energy sources, of which the share is still rather low in Poland, might constitute a long-term threat to conventional generation (see section 20.2.3.2, last paragraph). 36 Those calculations are based on electricity sold bilaterally (by generators and trading companies to other trading companies) and through power exchange. 37 Average prices in 2014. Ernst & Young, presentation (29 October 2014, Warsaw, n 69). 38 39 At n 45. The latest in 2012 (n 13). 40 The Energy Law reform was initiated in December 2011. On the one hand, it was necessary to implement the Third Energy Package (in particular 2009 Electricity Directive (n 24)), as well as Directive 2009/28 on the promotion of the use of energy from renewable sources and amending and subsequently repealing Directives 2001/77/EC and 2003/30/EC [2009] OJ L 140 (2009 RES Directive). On the other hand, the growing complexity of energy legislation called for separate legal regimes for electricity, gas, and RES. The government proposed a package of three acts (the new Energy Law for Electricity and Heating, the Gas Law, and the Act on RES), so that at least the three main branches of the energy sector could be regulated by the separate pieces of legislation (the so-called ‘energy three-pack’). The government’s draft proposal provoked a heated public debate, in particular regarding the scope and the level of RES support schemes in the Act on RES. Various lobby groups were exercising pressure and the proposed three-pack stalled by a bitter political deadlock. In order to avoid proceedings before the ECJ and heavy penalties (see IP/12/1139 of 24 October 2012; IP/12/1236 of 21 November 2012, IP/13/259 of 21 March 2013), the most urgent provisions in implementing EU directives have been processed independently from the three-pack and adopted in July 2013, in the form of just another amendment to the existing Energy Law Act (The Act of 26 July 2013 amending the Energy Law Act and related acts, Journal of Laws of 27 August 2013, item 984). This amendment has been labelled the ‘small’ energy three-pack, whereas legislative work on the initial ‘big’ three-pack has been dropped, with the exception of the Act on RES. The latter was finally adopted on 20 February 2015, and entered into force on 4 May 2015. The Act introduces an auction-based support scheme for the new RES, which would apply from 2016 in parallel to green certificates for the existing RES. The adopted Act has been heavily criticized and its amendment is already under consideration. 41 For instance, TAURON was planning to build a new 450 MW gas-fired unit at the Łagisza power plant together with PGNiG. In August 2013, PGNiG withdrew from the project due to uncertainty regarding support for cogeneration.

340

Poland

20.2.3.1 Ministry’s Report of 2013 and follow-up actions The capacity debate was further ignited in 2013, when the Ministry of Economy (and earlier the TSO) warned of possible capacity shortages in 2015–2019, due to the reasons outlined earlier (lack of investments coupled with plant decommissioning). According to the Ministry’s Report,42 12.26 GW of generation capacity would be decommissioned by 2030, and 4.4 GW of this in 2014–2017. Regarding new investments, the Ministry forecasted that, in total, 27.24 GW of new capacity should be developed in 2013–2030, almost half of which should already be in place by 2020. However, no significant investments were realized in 2013–2014, and until the end of 2016, capacity will remain at the 2012 level (approx. 38 GW), as a result of decommissioning and simultaneous increases in renewable and gas-fired generation.43 Based on this, the Ministry predicted that capacity shortages might begin in winter 2015, reaching a maximum of 1.1 GW in winter 2017, but should cease in 2019, as 6.7 GW of new hard coal, lignite, and gas-fired capacity is forecasted to come onstream between 2015 and 2019.44 In order to hedge against potential capacity deficits, the TSO introduced a new system service of cold intervention reserve. Furthermore, it modified the existing system service of operating capacity reserve. These capacity measures are considered interim solutions, aimed to ensure security of supply in the short to medium term (three to four years). They are described in more detail in sections 20.3.1.1 and 20.3.1.2 below. Additionally, the government (as majority shareholder) compelled the energy groups to continue large investment projects, which are not viable in economic terms. This includes the three biggest projects in coal-fired generation: 1.8 GW of new capacity at PGE’s power plant in Opole, 1 GW of new capacity at ENEA’s power plant in Kozienice, and 0.9 GW of new capacity at TAURON’s power plant in Jaworzno.45 Lastly, PSE entered into DSR contracts with large industrial users, as explained in section 20.3.1.3.

42

Ministry’s Report (n 18). In this Report, the Ministry also refers to the forecasts of PSE published earlier that year. In its report, submitted in March 2013 to the Parliament’s Energy Subcommittee, PSE analysed two scenarios. Under the realistic scenario, 6.6 GW of capacity would be decommissioned, whereas only 4 GW of new capacity would come on stream by 2020. As a result, PSE warned that Poland might face a capacity gap of 1–2 MW by 2016. Under the optimistic scenario, more capacity would come on stream by 2020 (6.5 GW instead of 4.4 GW), which might result in a short-term capacity problem in years 2016–2017. 44 Ministry’s Report (n 18) estimates that capacity shortage might occur not only in winter, but also in summer peaks. Deficits in winter peaks, ranging between 95 MW–1100 MW, could start as early as winter 2015 (95 MW). They would increase to 800 MW in winter 2016, to reach as much as 1100 MW during peak periods in the winter of 2017. As for summer peaks, the capacity gap is estimated to be between 30 MW– 680 MW. The first shortages might occur in summer 2016 with 520 MW, reach 680 MW in 2017, then drop to 30 MW in summer 2018. 45 For instance, in April 2013 PGE withdrew from the investment in Opole. In the words of the then PGE’s CEO, Krzysztof Kilian, ‘the project would make sense only if electricity prices were 50% higher’. PGE’s decision met with fierce opposition from the government, and open criticism from Donald Tusk, the Polish Prime Minister of the time. To minimize the investment risk, PGE entered into a long-term contract with Kompania Węglowa, a large state-owned mining company, for coal supply to the new units in Opole. Soon after, Krzysztof Kilian resigned from the post. 43

20.2 Setting the scene

341

20.2.3.2 URE’s Investment Report of 2014 The latest study of generation adequacy comes from the energy regulator, and is based on the most recent data provided by generators with more than 50 MW of installed capacity regarding their investment plans for the next fifteen years, as well as data on scheduled plant decommissioning and refurbishments. The expected changes in the country’s generation capacity, together with PSE’s forecast demand, are presented in URE’s Investment Report covering 2014–2028.46 The Report shows that the number of planned closures lowered as a result of the capacity measures taken by the TSO, and consequently the adequacy situation improved. Moreover, PSE forecasts a lower peak demand when compared to its earlier prognosis, and this new data used by URE also contributed to a more optimistic adequacy outcome.47 With respect to new investments, URE estimates that more than 18 GW of new capacity will come on stream in 2014–2028. Out of this, approx. 7.5 GW are investments in wind generation.48 Most of these new investments will be realized in years 2017–2019. It must be noted that URE’s Investment Report does not take into account potential investment in nuclear generation, which explains the large discrepancy between the figures, when compared with the Ministry’s Report (27.24 GW in 2012–2030).49, 50 Some additional capacity will result from the refurbishment of existing plants (approx. 350 MW), most of this in 2014–2017. In reference to plant closures, URE estimates that 5.2 GW of capacity will need to be decommissioned in the analysed period,51 less than forecasted in the Ministry’s Report and URE’s earlier estimations. Despite the observed improvement in the adequacy situation, URE still indicates periods of capacity reserve deficits. In particular, there is a risk of shortages during the two consecutive winters of 2015–2016, amounting to 1.2 GW and 1 GW respectively.52 However, URE admits that these potential capacity reserve deficits should be covered by generators which have not been included in URE’s study (generators with less than 50 MW of installed capacity). This additional, unreported capacity, together with emergency imports53 and the TSO’s capacity measures described below in section 46

47 URE’s Investment Report (n 23). URE’s Investment Report (n 23) p 18. However, if one only considers projects in the late stages of development (for instance, with necessary permits and financing, or after the tender, or those already in the construction phase), then the share of investments in wind decreases from 40% to 0.8%, whereas most of these ‘advanced’ projects relate to coal generation (60%). 49 The Polish Nuclear Power Programme, adopted on 28 January 2014, aims to pave the way for the construction of two nuclear power plants in Poland by 2025. The project is at the preliminary stage, with preparatory work being carried out jointly by a consortium of energy companies (PGE, ENEA, TAURON, and KGHM). Given the huge investment cost, it is still too early to judge whether the project will actually go ahead. According to the agreement between the investors, the decision will be taken in 2017 and will depend on whether or not the government develops a scheme to support nuclear generation. 50 URE also notes that this number (18 GW) is much lower when compared to generators’ investment plans in 2011, when the URE conducted its previous investment study (29.5 GW). 51 For more details, see text at n 30. 52 URE’s Investment Report (n 23) p 18. 53 Emergency import (or emergency supply) is an ancillary service which allows PSE to ask for assistance from the neighbouring TSOs, in order to cover a shortage of power within the Polish electricity system (or take the surplus power out of the system). This service can be activated if an unexpected operational situation occurs such as interconnector overloading, power shortage/surplus within the power system, nonfulfilment of the N-1 criterion etc. This service is reciprocal and activation can also be requested by a 48

342

Poland

20.3.1, would amount to 3 GW of available capacity in winter periods and up to 6 GW in summer periods. This is sufficient, according to URE, to cover peak demand and ensure the required level of operating reserve (9 per cent of forecast demand).54 Lastly, URE’s Investment Report mentions the increasing relevance of flexible resources in the system. It observes that wind generation constitutes more than 40 per cent of planned investments, whereas flexible gas-fired generation amounts to 22 per cent.55 In particular, flexibility might be needed in northern Poland, where most of the wind generation will be connected to the grid.

20.3 Energy-only market, capacity measures, and a ‘capacity debate’ Poland’s market is energy-only, but PSE took certain capacity measures, in 2013–2014, in order to ensure sufficient capacity reserves by keeping inefficient units in operation.56 These include the cold intervention reserve and the operating capacity reserve (see sections 20.3.1.1 and 20.3.1.2). Moreover, in 2013 PSE introduced demand side response (DSR) contracts (section 20.3.1.3). In parallel, there are proposals to introduce a market-wide capacity mechanism (section 20.3.2).

20.3.1 Capacity measures 20.3.1.1 Cold intervention reserve Cold intervention reserves are units kept by generators in stand-by, in order to provide electricity, at the TSO’s request, in case of a capacity shortfall. The mechanism targets old power plants, which are supposed to be decommissioned by 2016 due to environmental, technical, or economic reasons. This includes units, for which operation under the IED standards would be too costly, but which can enjoy the ‘limited life time derogation’, provided they run less than 17,500 hours during the derogation period (Article 33 of the IED). In that way, the oldest and least cost-effective units can be excluded from the market, but remain at the TSO’s disposal, and are only used for the balancing of the system. Providers of cold reserves receive fixed payments from the TSO for keeping their units in stand-by. If they are called to produce, the TSO also covers their generation costs (non-profit). The cost of cold intervention reserves is passed on to end consumers in the form of higher transmission tariffs.

neighbouring TSO with whom the PSE signed an agreement on emergency assistance. According to the PSE, the emergency assistance service via SwePol Link can provide for approx. 300 MW. 54 URE’s Investment Report (n 23) p 18. See also n 27. 55 URE’s Investment Report (n 23) p 19. However, almost all of investment projects in wind generation are in the early development stages, and the probability that they will be realized is lower when compared to more advanced projects in gas-fired generation. See also n 48. 56 It is estimated that the introduced capacity measures would prevent a shutdown of approx. 2 GW of generation capacity, in total (unprofitable units, old units or those which do not comply with stricter emission standards).

20.3 Energy-only market

343

The introduction of the cold intervention reserve did not require any legislative changes. So far, PSE organized two tenders for the provider of the cold reserve, and awarded two-year contracts to PGE and TAURON for 830 MW capacity, in total.57 The service will start in 2016, when IED renders the contracted units obsolete for regular use, and can be extended for two more years (until the end of 2019). Under these contracts, the TSO pays approx. 5 €/MW (24 PLN/MW gross) for every hour of keeping the units in stand-by.

20.3.1.2 Changes in the procurement of the operating capacity reserve According to the Grid Code, the TSO must ensure that the system has a sufficient level of capacity, not only to cover the projected peak demand, but also to allow for exceeding it by a set margin. This surplus of capacity, amounting to at least 9 per cent of forecast demand, and kept in the system for security of supply reasons, is called the operating capacity reserve. This type of reserve can only be provided by certain generating units, the so-called Power Generating Scheduling Units (JGWa). Contrary to the providers of the cold reserve, the JGWa providing an operating reserve can still participate in the market. The operating capacity reserve is thus a system service provided by certain eligible generators to the TSO and regulated by the Grid Code.58 In December 2013, the TSO modified the rules for the determination and settlement of this reserve in order to keep certain coal-fired generation units in operation, even though keeping them onstream is no more profitable.59 Based on these new rules, the TSO pays capacity payments to the generators for their availability, helping them in this way to cover their fixed costs. The payments are based on the reference price of the hourly operational reserve, which corresponds to the average unit technical fixed cost of the generating capacity of a given Power Generating Scheduling Unit, excluding depreciation and the expenses of the administration and sale, adjusted by an efficiency coefficient at the level of 0.93. For 2014, the value of the reference price is approx 8.6 €/ MWh (37.13 PLN/MWh).60 In 2014, capacity payments amounted to more than €93 million (PLN 400 million).61 This cost has been passed on to end consumers via the 57 As a result of the first tender organized at the end of 2013, the contract was awarded to two units of PGE’s Dolna Odra power plant (approx 454 MW). This was still not sufficient, according to the PSE, to effectively balance the system. Therefore, the PSE organized another tender in January 2014, which resulted in an additional 376 MW of the cold reserve capacity from TAURON’s power plants in Siersza and Stalowa Wola. See PSE, Press release of 26 March 2014, Konstancin—Jeziorna. 58 IRiESP (n 27). 59 PSE, Karta Aktualizacji nr CB/9/2013, IRiESP—Balancing, approved by the URE (Decision of the President of the URE of 10 December 2013 (DRR-4320-2(27)/2010/2013/JRz)). 60 The cost of the operating capacity reserve in 2014 exceeded the estimated cost (PLN 400 million/year) by 88 million. Therefore, the PSE considers lowering the reference price to PLN 17 in 2015. 61 According to the press, TAURON received € 35 million (PLN 153 million) for the provision of operating reserve in 2014. Other large providers include PGE (approx €4 million), ENERGA (€2.5 million), ENEA (€583.000). The TSO’s budget for an operating reserve in 2014 was exceeded by 22% and the TSO is going to decrease the reference price to 4 €/MWh (17 PLN/MWh) in 2015, subject to the URE’s approval. Press release of 8 September 2014, available at http://wysokienapiecie.pl/sieci/481-energetycy-walcza-orezerwe, accessed 1 February 2015.

344

Poland

so-called ‘qualitative coefficient’ (stawka jakościowa) included in the transmission tariff, approved by URE.

20.3.1.3 Tenders for DSR contracts Demand Side Response contracts is a voluntary scheme introduced in March 2013, whereby large industrial consumers commit to interrupt their electricity consumption, at the request of the TSO during peak demand, in exchange for a payment. The contract determines the volume of the reduction (minimum 10 MW), interruption time (two, three, or four hours) and the period in which the interruption can occur (winter or summer months). The TSO is entitled to request up to fifteen reductions per year, but not more than one reduction in a twenty-four hour period and not more than three reductions per week. Under this scheme, consumers are paid for each MWh of ‘saved’ electricity, but not for the availability to reduce demand. The payments vary depending on the contract.62 Until the end of 2014, PSE had organized five tenders for DSR contracts. Following the first four tenders, PSE signed contracts for a total capacity of approx. 150 MW. The fifth tender is still open63 and is expected to add an additional 200 MW of ‘saved’ capacity.64

20.3.2 Discussion on a market-wide capacity mechanism Following the Ministry’s Report in 2013, there has been a significant debate concerning the possible adoption of some sort of market-wide capacity mechanism, which could ensure generation investment beyond the 2020 horizon.65 However, in 2014, with capacity measures in place and major investment projects going ahead, the urgency of developing a capacity mechanism has dissipated for both the government and the regulator.66 Instead, unsurprisingly, the idea was pushed forward by the industry.

62 For instance, under the first contract signed by the TSO in March 2013, PGE (large industry consumer) will reduce its consumption when called upon by the TSO, and each ‘saved’ MWh will cost approx. €178.4 (PLN 750). It is argued that introducing such an additional payment for readiness to reduce consumption would make the scheme more attractive. 63 At the time of writing this chapter (January 2015). 64 Just to compare, 1 MWh of electricity which was not supplied due to the 2008 blackout in Szczecin cost PSE approx €3,100 EUR (PLN 13,000). 65 Forbes, Press release of 10 October 2013, available at http://www.forbes.pl/ministerstwo-gospodarki-dokonca-roku-projekt-rynku-mocy,artykuly,164410,1,1.html, accessed 1 February 2015. See also comments by URE in CIRE, Press release of 23 August 2013, available at http://www.cire.pl/item,80491,1,0,0,0,0,0,uremodel-rynku-mocy-powinien-byc-gotowy-w-ciagu-roku-juz-od-14-operacyjna-rezerwa-mocy.html, accessed 1 February 2015. 66 See, for instance, comments by the President of the URE, Maciej Bando, during a conference Capacity Market—a solution for Poland (Rynek Mocy—Rozwiązanie dla Polski), organized by the Polish Committee of the World Energy Council, 29 October 2014, Ministry of Economy, Warsaw. Based on CIRE, Press release of 30 October 2014.

20.3 Energy-only market

345

At the beginning of 2014, energy companies,67 together with the TSO,68 commissioned Ernst & Young to develop a capacity mechanism for the Polish electricity market (the industry proposal). The proposal was submitted to the Ministry and the energy regulator in November 2014, and could potentially constitute a basis for further legislative proposals from the Ministry. While the E&Y report is not publicly accessible, key features of the proposed mechanism were presented in October 2014 during a conference on capacity mechanisms hosted by the Ministry.69 The industry proposal analyses two possible types of a capacity mechanism—a centralized capacity mechanism (the UK model)70 and a decentralized market (the French model),71 whereby the centralized model is easier to implement, and could be adopted as soon as 2016. Both variants will be supported by contracts for difference (CfDs). According to the authors of the proposal, while a capacity mechanism would keep existing plants in operation covering their fixed costs, and thus ensure security of supply in the medium term, it might not create sufficient investment signals. Hence, CfDs are considered necessary to actually trigger investments in new units, in particular high-cost nuclear generation, and ensure security of supply in the long term.72

20.3.2.1 Centralized capacity mechanism (vs the UK model) The centralized model draws on the UK capacity auctions, which is discussed in depth in chapter 22. However, there are differences, as the model needs to take into account the specifics of the Polish market. First, while, in the UK, all licensed generators must apply to pre-qualify their eligible units (or submit an opt-out notification),73 in Poland only the so-called Centrally Dispatched Generation Units74 would be obliged to do this. Other generators could pre-qualify their units and participate in the auction on a voluntary basis. Secondly, instead of an auction in a descending clock format, it would be a sealed bid format in which providers state the minimum price they need and the auction is completed in a single round. However, capacity would be split in tranches, that is, generators would not submit plant-specific bids. Thirdly, following the UK model, auctions would be open to both new and existing power plants, although with different contract lengths. Contracts would be one-year-long with the possibility of an extension of up to ten years for new generation, five years for power plants requiring refurbishment and three years for other units.75 Lastly, determining the target volume of capacity to be procured would differ from the UK’s enduring reliability standard. In 67 Towarzystwo Gospodarcze Polskie Elektrownie (TGPE), Polskie Towarzystwo Elektrociepłowni Zawodowych, PGE. 68 The TSO supports the adoption of a capacity mechanism in the long term, indicating that generation adequacy problems will re-emerge after 2023, once the IED derogations expire. See comments of Henryk Majchrzak, CEO of the TSO, during the conference on the capacity market (n 66). 69 Conference on the capacity market (n 66). 70 71 See chapter 22. See chapter 14. 72 Conference on the capacity market (n 66), presentation of Stanisław Poręba and Maciej Przybylski (Ernst & Young), Rynek mocy—model dla Polski. Wprowadzenie do panelu dyskusyjnego. 73 See section 22.3.3.2. 74 Jednostki Wytwórcze Centralnie Dysponowane (JWCD) are units which are connected directly to the transmission network and are centrally managed by the TSO. 75 For contract lengths in the UK model, see chapter 22.

346

Poland

the proposed model, the TSO would calculate, based on the demand forecast, the volume of capacity needed to cover the forecasted peak demand and the required capacity reserve margin, subject to approval by the Ministry of Economy. Capacity payments would be financed through the so called ‘capacity charge’ included in the tariff. It is estimated that the mechanism would cost between 35,000–58,000 €/MW/year (150,000–250,000 PLN/MW/year). Auctions would be open to cross-border generators if they can demonstrate that they are physically able to deliver electricity.

20.3.2.2 Decentralized capacity mechanism (vs the French model) An alternative to centralized capacity auctions would be the French model of capacity obligations, discussed in chapter 14, with minor differences. These include longer contracts (three years instead of one), which would stabilize the market and mitigate the effects of price swings. Further, the suppliers’ capacity requirements do not need to be as precisely defined as they are in the French capacity mechanism.76 Contrary to France, growing peak demand is less of a problem in Poland, so the suppliers’ capacity requirement can be determined simply through their contracted volumes. Lastly, the authors of the proposal suggest that the secondary market for capacity certificates can draw on the Polish experience with other similar support mechanisms currently in place. One of the advantages of the decentralized model (over the centralized one) is the fact that capacity payments are included directly in the energy prices, and not as a separate ‘capacity charge’, fostering price competition between energy suppliers. However, the industry considers the decentralized model a second-best option, given its more complex implementation. Also, since the Commission formally approved the UK capacity mechanism,77 the UK model provides some guidance to a state aid conforming capacity auction.78

20.3.2.3 Contracts for Difference (CfDs) Alongside the capacity mechanism, the industry proposes to introduce Contracts for Difference, a solution recently implemented in the UK to support RES and nuclear generation, and approved by the Commission under state aid rules.79 Contracts would be concluded between generators and a state-owned settlement body created for that purpose.80 Under CfDs generators receive a guarantee of a fixed price for their electricity sold on the market, instead of being exposed to fluctuating wholesale prices (strike price). If the wholesale price is lower than the strike price determined in the contracts, the generator receives compensation to match the strike 76

Compare with section 14.3.2.1. Commission decision of 23 July 2014 in Case SA.35980 (2014/N-2) United Kingdom Electricity Market Reform—Capacity Market C (2014) 5083 final, [2014] OJ C/348 (UK capacity mechanism). 78 See section 9.5.1 for an assessment of the Commission’s decision. 79 See chapter 22. 80 ‘Zarządca Rozliczeń Różnicowych SA’, incorporated as a special purpose public company. 77

20.4 European dimension

347

price. To the contrary, if the wholesale price is higher than the strike price, generators would be required to pay back the excess of revenue above the strike price. CfDs provide generators with greater predictability of revenue and financial stability in the long term, encouraging them to invest. Thus, for each generator, the strike price must be set at a level ensuring the profitability of investment, given the chosen technology and other incomes (eg from other support schemes). It is assumed that CfDs would be for twenty years, which is longer than currently observed in the UK (fifteen years). In the case of nuclear generation, contracts would be extended to forty years.

20.3.2.4 Timeline The industry proposal assumes that a capacity mechanism could already be in force in 2016, following a very ambitious implementation schedule. This includes a formal proposal from the Ministry before the end of 2014 (already overdue) and all the necessary legislative amendments in place, together with the State aid clearance from the Commission, by November 2015.

20.4 European dimension The Ministry admits that the adoption of a capacity mechanism would require negotiations with the Commission and would be subject to the Commission’s approval.81 Thus, any solutions set out in the industry proposal have been assessed with respect to compliance with EU law, in particular EU state aid rules.82 This explains why the industry proposal heavily relies on the UK and French market models, the only capacity mechanisms so far which have been negotiated and approved by the Commission.83 The following paragraphs discuss the centralized capacity mechanism (the preferred model) from the perspective of EU state aid control. A centralized capacity mechanism based on the UK model of capacity auctions would, most probably, be considered state aid under Article 107(1) TFEU, and would need to be assessed under the Commission’s new EEAG for the years 2014–2020.84 According to these Guidelines, any form of aid compatible with the common market should contribute to an objective of common interest (in this context, generation adequacy), and Member States are discouraged from solving adequacy problems by supporting environmentally or economically harmful generation. With regard to that, the proposed centralized capacity auctions would be open to both existing and new power plants, following the UK model. Moreover, it appears that the objective of the proposed auctions is to keep the existing plants in operation, rather than incentivize 81

Conference on the capacity market (n 66). Poręba and Przybylski (n 72). The report including this assessment is intended for internal consultation and is not publicly available. 83 UK capacity mechanism (n 77). The French capacity mechanism was not considered State aid in the first place, but the French government agreed all the proposed solutions with the Commission, and the mechanism was approved. See chapter 14. 84 Communication from the Commission, Guidelines on State aid for environmental protection and energy 2014–2020 [2014] OJ C 200/1–55 (EEAG 2014–2020). 82

348

Poland

investments, as the latter would be ensured by CfDs.85 Therefore, allowing the existing old coal-fired units to bid in the auction entails the risk of subsidizing them, instead of more flexible gas-fired units. In that respect, it is useful to look at the results of the first UK capacity auction from December 2014, where the capacity price (19.4 £/kW) was apparently too low for new investment projects, pushing them out of the auction despite the prospect of longer contracts.86 The assumption that the Commission would approve of the Polish capacity mechanism given its similarities to the UK scheme, may be unfounded. Every support scheme is assessed on a case-by-case basis, and factors such as Poland’s dependency on coal (87 per cent) compared to the UK (39 per cent) may prove to be significant differences in securing approval. Despite the inconclusive wording in the EEAG, which only encourages (and not obliges) Member States to avoid environmentally harmful subsidies, it would be most welcome if the old and high-emitting coal-fired power plants were excluded from the auction upfront, for instance, by devising strict eligibility criteria. Furthermore, under EEAG, Member States need to consider alternative solutions for capacity problems, before they adopt a capacity mechanism. Recent changes to the provision of reserve capacity, tenders for DSR contracts, the planned installation of phase shifting transformers on the Polish–German interconnections,87 as well as building new interconnectors (eg the power interconnection Poland–Lithuania)88 can be considered as alternative measures to employing a market-wide capacity mechanism. Should the government renew interest in developing a market-wide capacity mechanism, it would have to demonstrate the insufficiency of the adopted measures to avert the risk of adequacy problems in mid-2020. Another question is whether the new rules for the provision of the operating capacity reserve could be considered state aid and, as such, need to be submitted to the Commission for approval. The amount of capacity payments awarded based on this mechanism in 2014 (for instance, €35 million for TAURON) qualifies the measure for the assessment under the EEAG (more than €15 million per project, per undertaking). The Commission has not been notified of the amendment of the Grid Code, which triggered these payments, as the government would likely argue that the payments have been awarded in exchange of PSOs.89 However, in order to be considered a compensation for PSOs, the service of providing an operating reserve would need to fulfil the four Altmark conditions. It might be questioned whether the current rules are transparent and precise enough to constitute a validly imposed PSO. It is interesting to note that both the energy regulator and the TSO have indeed admitted that the current rules governing the provision of the operating capacity reserve are not clear and need to be further amended. Moreover, since the generators are not selected through an open tender

85

Poręba and Przybylski (n 72). National Grid, Provisional Auction Results, T-4 Capacity Market Auction 2014, available at https://www.gov.uk/government/uploads/system/uploads/attachment_data/file/389832/Provisional_Results_ Report-Ammendment.pdf, accessed 1 February 2015. 87 URE, National Report (n 3) p 22. 88 See www.polaczeniepolskalitwa.pl/polaczenie-polska-litwa-o-projekcie.php, accessed 1 February 2015. 89 PSOs are public service obligations. See section 9.3.3 for a discussion. See also section 1.3.2 on the Irish CADA case. 86

20.5 Conclusions

349

procedure, it is doubtful whether the service would meet the fourth condition of the Altmark test. Here, the government would need to show that the awarded payments have been determined in relation to the costs of a typical, well-run undertaking, adequately financed to meet the requirements of public service. As explained above, in section 9.3.3, this is not always easy for the Member State to demonstrate. The Commission attaches particular importance to the participation of demand side response solutions and cross-border interconnectors in the capacity mechanism. With that in mind, the industry proposal suggests that capacity auctions should be open not only to demand side response solutions, but also to generators located abroad, if they can prove their physical ability to deliver electricity. In other words, they would need to show that they dispose of sufficient cross-border capacity. This condition de facto excludes them from the auction, as Poland simply does not have enough cross-border capacity.90 Further, even if there was sufficient capacity on the interconnectors, it would be extremely difficult to enforce the foreign generator’s commitment to deliver its electricity.91 As suggested by the decision on the UK capacity mechanism,92 although the mechanism excludes generators from the neighbouring countries, it can still be cleared by the Commission under EEAG, if the Member State proves that including them is physically not possible, either because there is insufficient crossborder capacity, or the state would not be able to sanction foreign generators who fail to deliver capacity. Both claims can be raised by the Polish government. Lastly, a parallel can be drawn between CfDs and the long-term power purchase agreements (PPAs), a mechanism which existed in Poland before market liberalization. In 1993–1998 the Polish grid company entered into long-term PPAs with the generators providing them with a guarantee of purchase of specified volumes of electricity at predefined prices. The longest agreements were supposed to expire in 2020. In total, they covered about 50 per cent of the country’s electricity trade, aimed at securing bank loans for the generators’ investment projects. In 2007, the Commission requested that Poland terminate all PPAs, as they constituted unlawful and incompatible state aid.93 At the same time, the Commission authorized a temporary support scheme to compensate generators’ stranded costs, resulting from the forced termination of their agreements.94 The compensation is still financed from the so-called ‘temporary charge’ (opłata przejściowa) included in the tariffs, and will expire in 2020. This explains why the industry proposes to introduce CfDs, aimed at securing financing for investments beyond 2020.

20.5 Conclusions It is still too soon to speculate if, or when, Poland would adopt a market-wide capacity mechanism. Energy groups have advocated and pressured for the introduction of a 90 In theory, cross-border capacity constitutes approx 10% of Poland’s projected peak demand (both import and export). In practice, it amounts to 400 MW in the case of exports (and subject to temporary network congestion), and 0 MW in the case of imports, based on PSE’s data for 2015. 91 Chapters 3 and 6 deal with this problem in more detail. 92 UK capacity mechanism (n 77). See section 9.5 for an assessment. 93 European Commission, Press release IP/07/1408 of 27 September 2007. 94 For stranded costs, see section 9.2.

350

Poland

capacity mechanism based on the UK model, with the first auctions already expected to take place in 2016. While the Ministry expresses its support for a long-term solution to adequacy problems, and actively engages in the discussion with the industry, it is doubtful that any legislative proposals will be considered before the parliamentary elections in September 2015. The risk of capacity shortages in the period 2015–2019 has been mitigated by the adoption of short-term capacity measures. Accordingly, both the energy regulator and the TSO admit that the system should not face any capacity problems before 2023. This suggests that the government might delay market reforms until 2017 at the earliest, allowing it to observe the performance of the UK capacity mechanism.

21 Spain In˜ igo del Guayo

21.1 Introduction The aim of this chapter is to analyse capacity payments in Spain. It tries to explain why they were introduced in 1997, at the time of liberalization, and how they have developed until the recent entry into force of a new Electricity Act in December 2013. Section 21.2 describes the Spanish generation mix, the country’s electricity regulation, generation adequacy, and the procedure for the conferral of generation authorizations, with a special reference to the so-called ‘electricity deficit’ which lies behind the recent reform of the Spanish capacity mechanism. Section 21.3 focuses on capacity payments. After explaining the regulatory context and the key features of the mechanism, the section discusses the reasons for the reductions in payments (and their abolishment in certain cases) between 2012 and 2013. Moreover, it illustrates the Spanish Supreme Court’s approach to the reductions in capacity payments, as well as their regulation in the Spanish energy law. Section 21.4 highlights the relevance of the Spanish capacity payments in a broader European context. Finally, section 21.5 offers some concluding remarks.

21.2 Setting the scene 21.2.1 Market characteristics Spain is not an energy-only market, as generators receive capacity payments.1 Prior to liberalization, new generation capacity was constructed according to the government’s plans, which were compulsory for companies. Since the market liberalization in 1997, market-based investment decisions of generators have been supplemented with a system of capacity payments in order to, on the one hand, guarantee the construction of new capacity (if the market does not deliver it) and, on the other hand, to provide flexible back-up generation for intermittent RES. These payments were modified in 2007, and later reduced or abolished in 2012 and 2013. Electricity production in Spain is based on wind, nuclear, coal, hydro, co-generation, natural gas, solar photovoltaic, biomass, and solar thermal.2 There has been a remarkable increase in the use of natural gas for electricity production in the past twenty years, 1

See section 1.2.3.5 for key elements of this mechanism. Generation sources ranked by share in Spain’s electricity production in October 2013 (listed in accordance with its relevance): 21.1%—wind, 21%—nuclear, 14.6%—coal, 14.4%—hydro, 12.4%—cogeneration and other sources, 9.6%—natural gas, 3.1%—photovoltaics, 2%—biomass, 1.8% solar thermal. See REE, Monthly Bulletin No 82, October 2013. 2

352

Spain

as well as of renewables. Spain’s gas-fired generation exceeded coal-based generation for the first time at the end of 2005, becoming then the most important type of electricity generation. However, wind power generation has recently increased its share in Spain’s electricity generation and overtook other fuel sources in 2013. The cumulative annual electricity demand in 2013 amounted to 246,314 GWh (2.2 per cent less than in 2012). Only a small amount of electricity is traded across the borders, particularly with France. Spain’s annual cross-border trade balance is often negative, ie exports from Spain exceed imports. In 2013, balance with France was positive, whereas balances with Portugal and Morocco were slightly negative.3 The limited export volumes from a country with excess generation capacity demonstrate the isolation of a peninsular electricity system, poorly connected to the rest of Europe. The problem of sufficient interconnection capacity should diminish in the future, with the new electricity line across the Pyrenees.4

21.2.2 Regulatory framework Liberalization of the electricity market in 1997 was introduced by the Electricity Sector Act (ESA 1997),5 which set up an hourly electricity spot market (electricity pool).6 All generating companies can trade bilaterally with buyers for the delivery of electricity, but they are obliged to sell their remaining output (not covered by bilateral contracts) to the pool at the pool’s clearing price. The pool’s price is determined by matching offers from generators to bids from consumers in each hour and equals the short-run marginal cost of the plant generating the last unit of electricity required to meet demand in that hour (marginal pricing model).7 The ESA 1997 has been repealed and substituted by the 2013 Electricity Sector Act (ESA 2013).8 The Spanish electricity industry consists, basically, of five private and vertically integrated electricity groups, which produce, distribute, and supply electricity: Iberdrola (47.3 per cent), Endesa (owned by the Italian Enel, 27.7 per cent), Gas Natural Fenosa (14.6 per cent), HC Energía (owned by the Portuguese EDP, 5.8 per cent), and E.ON (2.8 per cent).9 3

See REE, Monthly Bulletin No 84, December 2013. See Decision 1364/2006/EC of the European Parliament and of the Council of 6 September 2006 laying down guidelines for trans-European energy networks and repealing Decision 96/391/EC and Decision 1229/2003/EC [2006] OJ L 262/1, which includes a new trans-Pyrenean interconnection between France and Spain (see Annex III). 5 Electricity Sector Act No 54 of 27 November 1997, Spanish Official Bulletin No 285 of 28 November 1997 (ESA 1997). 6 An electricity pool or electricity market is an organized market where players (producers and consumers) settle their transactions on energy. A pool can be either voluntary or compulsory, depending on where electricity generated can or must (respectively) be sold and bought. The price of electricity is formed in the pool by means of the free play of supply (electricity generators) and demand (wholesale sale or final consumers). The pool model is explained in sections 1.1 and 2.2 above. 7 For further information on the Spanish pool, see http://www.OMIE.es, accessed 1 February 2015. 8 Electricity Sector Act No 24 of 26 December 2013, Spanish Official Bulletin No 310 of 27 December 2013 (ESA 2013). 9 The remaining 1.8% being in the hands of other suppliers. See the CNMC, Report on the retail electricity market from June 2011 to June 2012, available at the CNMC’s website (http://www.cnmc.es, accessed 1 February 2015). 4

21.2 Setting the scene

353

Main actors responsible for generation adequacy are the Ministry for Industry and Energy, the National Commission for the Markets and Competition (CNMC),10 and the Spanish TSO (REE), which is responsible for operating the Spanish electricity transmission grid and ensuring optimal network planning, in order to coordinate electricity generating units with consumption points. In several indirect ways (such as country planning), the seventeen Spanish Autonomous Communities (for example, Catalonia or the Basque Country) influence decisions on investments in new generation capacity.11

21.2.3 Generation adequacy Generally, the problem of generation adequacy in Spain is presented as both an issue of security of supply and of flexibility. It is of course vital to ensure sufficient generation capacity to cover demand in the long term. This is the security of supply component. Generation adequacy is also a problem of flexibility. On the one hand, due to weather conditions, there is a remarkable peak demand in summer, coming from air conditioning needs. There are also peak demand periods in winter which contribute to generation adequacy problems, but they are not so frequent in Spain. On the other hand, some electricity generation, like RES, is variable and there is a need to have sufficient capacity ready to go onstream when there is no wind or no sun. The Spanish electricity system has so far not experienced generation adequacy problems (in the meaning described earlier), even during periods of peak demand. Rather, problems arose due to congestion in the Spanish transmission network, ie the lack of sufficient transmission capacity to transport electricity from generating units to areas of high electricity consumption. Transmission congestion in the Spanish grid led to a number of blackouts in 2001 and to changes in the rules related to congestion management and grid planning. This means that security of supply is not only an issue of generation adequacy, but also of transmission adequacy. Therefore, the proper fulfilment of security of supply obligations includes providing sufficient transmission capacity, and is acknowledged in all Spanish legal acts regulating capacity payments. In fact, the problem of Spain is not that of lack of generation capacity, but rather of an overcapacity, which is a result of poor capacity planning decisions of the preliberalization regime. Currently, due to a growing share of renewables in the system, some thermal units (in particular, gas-fired power plants) face reduced running hours, being only a back-up for cheaper but intermittent RES. This problem is further exacerbated by the economic crisis which has resulted in a significant reduction of 10 The former National Commission for Energy was abolished in 2013 and integrated within a regulatory body with broader competences, the National Commission for the Markets and Competition (NCMC), created by Act No 3 of 4 June 2013, Spanish Official Bulletin No 134 of 5 June 2013. 11 See further on the Spanish market structure and regulatory framework, Ernesto Bonafe, Towards a European Energy Policy. Resources and Constraints in EU Law (Germany: Lambert Academic Publishing, 2012), in particular Chapter I. See also Javier de Cendra de Larragán and Véronique Bruggeman, ‘Spain’s energy law’ in Blanpain (ed), International Encyclopaedia of Laws: Energy Law (The Netherlands: Kluwer Law International, 2007) pp 1–321; and Iñigo del Guayo, ‘Energy law in Spain’ in Roggenkamp, Redgwell, Del Guayo and Ronne (eds), Energy Law in Europe. National, EU and International Regulation, 2nd edn (Oxford: Oxford University Press, 2007) pp 1077–168 (the 3rd edn forthcoming 2015).

354

Spain

energy demand. Consequently, the recovery of thermal power plants’ fixed costs from the energy markets has been severely affected. On the top of that, construction of new generation capacity in Spain encounters regulatory barriers, resulting from the long and complicated procedures to build new networks. Consequently, the lack of capacity would not be considered to be a proper market failure, but rather a regulatory failure. The Spanish approach to generation adequacy is linked to security of supply. The ESA 2013 provides that the Spanish government and CNMC have to intervene in the electricity sector in order to guarantee sufficient electricity supply at all times.12 CNMC forecasts in its 2013 Report on demand coverage13 that the Spanish electricity demand should not change significantly over the next few years. However, the uncertainty linked to demand forecasting impacts both investment in new generation as well as investment in refurbishment and maintenance of existing power plants. In this context, the planned new generation capacity is very small and corresponds to new hydroelectricity plants.14 Construction of CCGTs is not foreseen. Some gas- and coal-fired units are expected to be decommissioned in the next two or three years, due to the lack of demand.15 Regarding demand coverage, CNMC forecasts no substantial changes to the current generation adequacy levels. It estimates that demand coverage in 2013–2016 will have a safety margin of 10 per cent both for winter and summer peaks.

21.2.4 The Spanish electricity tariff deficit Any debate on energy regulation in Spain is currently dominated by the overwhelming problem of the so-called ‘electricity tariff deficit’. Despite electricity market liberalization in 1997, the Spanish government kept regulated end-user tariffs for household consumers and small enterprises. Instead of taking a cost-oriented approach in setting these tariffs, the government has consistently kept their value artificially low, well below the market price. This resulted in a tariff deficit, that is, the difference between the actual cost of producing electricity and the regulated electricity prices which are too low to cover this cost. In legal terms, this deficit is a debt, that is, the ‘electricity system’ owes electricity companies the quantity which the deficit amounts to (approx €15 billion). As of 2011, the government has decided to reduce the debt by gradually increasing tariffs and by setting up a securitization vehicle (Fondo de Amortizacion del Deficit Electrico, FADE), which will fund the tariff deficit by issuing bonds aimed at reimbursing electricity companies. Among the costs of the electricity system not covered by tariffs are RES subsidies. In order to further reduce the deficit, Spanish governments from 2008 to 2012 have

12

ESA 2013 (n 8). CNMC (the then CNE), Informe Marco sobre la cobertura de la demanda de electricidad y gas para los próximos años, Ref No PA006/12, 17 April 2013, available at http://www.cne.es/cne/doc/prensa/NP_17_4_ 13.pdf, accessed 1 February 2015 (2013 Report on demand coverage). 14 Two are about to open (1,240 MW in total) and there are other six projects (2,060 MW in total). 15 A gas plant in Arcos de la Frontera in the South of Spain (800 MW), and a coal-bed methane plant in Puertollano, central Spain. 13

21.2 Setting the scene

355

reduced subsidies to existing renewable installations and have reduced subsidies to new installations. This deficit also resulted in recent government’s decisions to reduce or abolish some capacity payments (further discussed in section 21.3.4).

21.2.5 Authorization procedure for the construction of new generation capacity Under the 1997 ESA economic planning became indicative and remained compulsory only in the case of grid infrastructure investment. In drafting the electricity infrastructure indicative planning, there is the need to take into account, among others, demand forecast as well as the minimum installed capacity to meet the expected demand under the criteria of security of supply, energy diversification, energy efficiency and environmental protection.16 State control over generation is now limited to granting authorization, as explained next. In accordance with both the 1997 ESA and the 2013 ESA, construction, operation, substantial modification, and decommissioning of all electricity generation units is subject to a prior administrative authorization granted by the Spanish government, whenever the installed capacity is of 50 MW or above. For units below 50 MW, the authorization is given by the government of the Autonomous Community where the installation is to be located. The granting of the authorization is regulated (ie not discretionary) and governed by the principles of objectivity, transparency, and nondiscrimination. The procedure is open to all investors, irrespective of their origin. However, the authorization to construct new generating units can only be granted to those investors which have received prior authorization to connect their facilities to the Spanish electricity grid.17 Grid access is regulated in REE’s Operating Procedure which may restrict grid access in certain network nodes or zones.18 The Operating Procedure provides that any access restrictions should be resolved taking into account the lack of a reserve system of network capacity. Further, time precedence in the connection (‘first come, first served’ rule) does not involve a correlative preference in access. Granting of access to the grid is based on market mechanisms, as provided in the Operating Procedure and specific rules for RES and cogeneration units.19 Detailed rules on the grid access are provided in the Royal Decree 1955/ 2000.20 The Spanish law does not foresee, for the time being, any tender procedure for the construction of new generation capacity as envisaged in Article 12 of the 2009 Electricity Directive.21

16

17 ESA 2013 (n 8) Art 4. ESA 2013 (n 8) Art 53(3). REE, Operating Procedure 12.1 (Applications for the connection to new installations to the transmission network), approved by a Decision of the General Secretary for Energy of 11 February 2005, Spanish Official Bulletin No 51, 1 March 2005. 19 See Decision of the General Secretary for Energy (n 18) point 4.1. 20 Royal Decree No 1955 of 1 December 2000, Spanish Official Bulletin No 310 of 27 December 2000. 21 Directive 2009/72/EC of the European Parliament and of the Council of 13 July 2009 concerning common rules for the internal market in electricity and repealing Directive 2003/54/EC [2009] OJ L 211/55 (2009 Electricity Directive). 18

356

Spain

21.3 Capacity mechanism 21.3.1 Background Spain introduced capacity payments in 1997. The 1997 system was replaced in 2007 with a new system of capacity payments, which subsequently have been reduced or abolished by the 2013 ESA.22 ACER observes that Member States may pursue several different policy objectives with their capacity mechanisms.23 Spain’s capacity payments aim at (a) ensuring generation adequacy, that is sufficient generation capacity to meet demand at all times, (b) maintaining system flexibility, ie responsiveness to sudden demand variations or unexpected outages, (c) reducing investment risks in new generation and avoiding price volatility. The current structure of capacity payments addresses system flexibility issues and reduces investment risks. Electricity market liberalization in Spain in 1997 meant a departure from the central planning approach. Prior to liberalization, electricity companies (both private and public) were compelled to invest in new generation capacity in line with governmental plans and their remuneration was also determined by the state according to ‘standard costs’ established in advance. Security of energy supply was one of the government’s priorities, which resulted in inefficiencies and capacity surplus, since more capacity was constructed than the system actually required.

21.3.2 Capacity payments between 1997 and 2007 Capacity payments were introduced in Spain in 1997 by the Royal Decree 2019/199724 and further regulated in the Ministerial Order of 17 December 1998.25 Their legal basis can be found in the ESA 1997,26 which states that the revenue of electricity generators shall include a payment for the guarantee of capacity actually delivered to the system. It appears that the Spanish government did not consider any alternative capacity mechanisms at that time. It is also not clear why the government decided on this type of capacity mechanism. However, as observed in section 1.2.3.5, capacity payments are characteristic of markets that are less well connected in the periphery of Europe, and have been also introduced in Ireland, Portugal, Greece, and Italy. The main reason for introducing capacity payments in Spain was, on the one hand, to help generators cover their costs in a market with price regulation and, on the other hand, to compensate their stranded costs27 during the transition from regulated markets to an open market. 22

ESA 2013 (n 8). ACER, Report on capacity remuneration mechanisms and the internal market for electricity, 30 July 2013 (ACER’s Report) p 9. ACER’s view on capacity mechanisms is set out in chapter 2. 24 Royal Decree No 2019/1997 of 26 December 1997, Spanish Official Bulletin No 310 of 27 December 1997. 25 Ministerial Order of 17 December 1998 developing some aspects of Royal Decree No 2019/1997, Spanish Official Bulletin No 310 of 28 December 1998 (MO 1998). 26 ESA 1997 (n 5). 27 Stranded costs are those costs which electricity companies had incurred before liberalization, due to regulatory obligations, and which will not be recovered under the new liberalized framework. See chapter 9 for a further discussion. 23

21.3 Capacity mechanism

357

The level of the payment depended on fuel source and the amount of capacity guaranteed by a given generator both in the medium and long term, determining the capacity price with respect to the long-term needs of the electricity system.28

21.3.3 Capacity payments from 2007 to date The ESA 1997 was amended in July 2007 in order to implement the 2003 Electricity Directive.29 Since then, the ESA 1997 stated that revenues of generators may (instead of shall) include a capacity payment, based on the capacity needs of the national electricity system. The change in the wording of the ESA 1997 suggested that the system of capacity payments is not obligatory anymore, but gave no further indication as to how the new system of capacity payments will differ from the existing one. In other words, the 2007 amendment left it up to the Spanish government to decide whether to keep capacity payments or not. This was an attempt to bring the Spanish law in line with the provisions of the 2003 Electricity Directive which provided for transparency and nondiscrimination in the generation markets, which may not be ensured in a market with capacity payments. Details of the new system were provided a few months later in the Ministerial Order ITC/2794/2007 (MO 2794/2007).30 Despite the MO 2794/2007 primarily focusing on the review of electricity tariffs, the new system of capacity payments is set out in Annex III of the act. In the new system, capacity payments remunerate capacity in two cases. First, there are payments to incentive investment in new capacity by helping generators to recover their investment cost (these are payments for new projects only). Secondly, there are payments for the so-called ‘availability service’, which aim to secure capacity in the medium term (up to one year).31 The first novelty of the new system set up in 2007 is that capacity payments take into account transmission bottlenecks in the Spanish grid. Namely, the preamble of the MO 2794/2007 explains that capacity payments rely on three premises: (a) inelastic electricity demand, (b) significant transmission constraints in the Spanish grid and (c) electricity price, which is insufficient to ensure the coverage of electricity demand. These three premises explain why capacity (ie the availability of providing electricity) has the character of a ‘public good’, and why its remuneration should be regulated in order to ensure the proper balance between supply and demand in the medium and long term at all network nodes. 28

THEMA consulting group, Capacity mechanisms in individual markets within the IEM, Report for the Directorate-General Energy of the European Commission, June 2013, p 39. 29 Directive 2003/54/EC of the European Parliament and of the Council of 26 June 2003 concerning common rules for the internal market in electricity and repealing Directive 96/92/EC [2003] OJ L 176/37 (2003 Electricity Directive). 30 Ministerial Order No ITC/2794 of 27 September 2007, Spanish Official Bulletin No 234 of 29 September 2007 (MO 2794/2007). 31 Certain provisions of the MO 2794/2007 (n 30) were modified by the Ministerial Order No ITC/3860 of the 28 December 2007, Spanish Official Bulletin No 312 of 29 December 2007, reviewing tariffs as from 1 January 2008, and by the Ministerial Order No ITC/3801 of 26 December 2008, Spanish Official Bulletin No 15 of 31 December 2008, reviewing tariffs as from 1 January 2009.

358

Spain

The second most important change is that the MO 2794/2007 introduces a different method to determine the amount of the payments, both in case of payments for investments and for the availability service. In the former case, the MO 2794/2007 links the evolution of investment incentives (long-term availability) with the coverage ratio or index, helping to make accurate forecasts of future capacity requirements, to avoid discretionary fluctuation of payments. In the latter case, the MO 2794/2007 clearly determines periods in which capacity is required, and foresees high penalties for non-delivery, incentivizing compliance. Remuneration of investments is in the form of a capacity charge for new plants (a contracted price per MW for each plant), and was capped at 28,000 €/MW per year for the first ten years. However, the actual payment was to be decided by CNMC based on a capacity price curve, as a function of the reserve margin, in the first year of construction. In other words, CNMC sets the price of capacity and the market chooses its amount by entry.32 The availability service (in the medium term) allows REE to enter into one-year contracts (or shorter) with selected power plants based on technologies which, most likely, could not be dispatched during periods of peak demand, because, for example, regular operation in the energy market prevent those installations to recover the fixed costs, as might be the case of oil-fired power plants. Payments for the availability service are managed by REE based on the principles of transparency and efficiency. The MO 2794/2007 was substantially modified by the Ministerial Order ITC/3127/ 2011 (MO 3127/2011), both with respect to payments for availability service and payments for investments in new generation.33 The preamble of the MO 3127/2011 explains that the effects of the economic crisis in the Spanish energy sector have resulted in a major decrease in electricity demand. Moreover, Spain has committed to produce 20 per cent of primary energy from RES by 2020, which implies that the share of RES in the system will further increase at a significant rate. Lower electricity demand and the growing share of RES in the Spanish electricity system have had a significant impact on the revenues of electricity producers who are responsible for ensuring the balance between supply and demand in the medium and long term (gas-fired generation, in particular). These producers become unprofitable, and may exit the market. Further, lower electricity demand does not incentivize new investment in capacity in the long term. Moreover, there has been a significant reduction in the operating hours of some technologies, which could not be compensated with an increase in export, due to the lack of sufficient interconnection with the rest of the EU. In these circumstances, the government decided to limit capacity payments only to those units which were actually providing security to the system. This requires a proper definition of the ‘availability service’, meant as the availability in a one-year time horizon, preventing certain units to exit the market. Availability

32 THEMA consulting group (n 28). See further, Carlos Batlle et al, ‘Enhancing power supply adequacy in Spain: Migrating from capacity payments to reliability options’ (2007) Energy Policy 35 (9), 4545–54. 33 Ministerial Order No ITC/3127/2011 of 17 November 2011, Spanish Official Bulletin No 278 of 18 November 2011 (MO 3127/2011).

21.3 Capacity mechanism

359

payments are based on the net power of the plant, as well as an index of availability. It applies to oil-fired plants, CCGTs, and coal-fired generation. It is also applied to some hydroelectricity units (pumping and reservoirs). This, in short, is the scope of the MO 3127/2011.

21.3.4 The reduction and abolishment of some of the capacity payments in 2012 and 2013 Taking into account a remarkable low electricity demand, Royal Decree-Law 13/2012 (RDL 13/2012)34 reduced (exceptionally for 2012) the annual amount of capacity payments for investment in long-term capacity to €23,400 per MW per year (the initial amount being €28,000 per MW per year). The reduction applied only to those generators who were receiving remuneration for investment when the RDL 13/2012 came into force. By means of the Royal Decree-Law 9/2013,35 the system of capacity payments for the promotion of long-term investments was further modified. First, capacity payments were reduced by more than a half for existing installations. Secondly, the period in which capacity payments are to be paid was extended. Thirdly, capacity payments for new installations were abolished. The preamble of RDL 9/2013 explains the changes as follows: In the current context in which there is an intense reduction of electricity demand, and where there is minimal risk of capacity deficit, it is considered urgent to extend the reduction of the incentive [ . . . ], accompanying this reduction by an extension of the remaining time in which existing facilities entitled to the capacity payment, at the time of the entry into force of this RDL, will be getting said payment.

The first measure introduced by the RDL 9/2013 is the reduction in payment for the promotion of long-term investment from €23,400 per MW per year, set by RDL 13/2012 for 2012, to €10,000 per MW per year for 2013. The second measure introduced by the RDL 9/2013 extends the period within which current power installations will be entitled to receive the capacity payment from ten years (set by the MO 2794/2007)36 to twenty years. This is meant to compensate for the sharp reduction in payment. The MO 2794/2007 establishes that the owner of installations is eligible to receive capacity payment from the date the given installation is registered with the Administrative Register of Electricity Production Facilities, for a period of ten years. RDL 9/2013 states that those facilities which were entitled to receive capacity payments at the time of its entry into force will actually get the compensation within a period which is twice the time to cover the ten year period introduced by MO 2794/2007. 34 Royal Decree-Law No 13/2012 of 30 March 2012, Spanish Official Bulletin No 78 of 31 March 2012 (RDL 13/2012). 35 Royal Decree-Law No 9/2013 of 12 July 2012, Spanish Official Bulletin No 167 of 13 July 2013 (RDL 9/ 2013). 36 MO 2794/2007 (n 30).

360

Spain

Thirdly, RDL 9/2013 abolishes capacity payments for the promotion of long-term investments for new production facilities, unless they get the administrative certificate enabling them to start operation prior to 1 January 2016, in which case they are entitled to €10,000 per MW per year for twenty years. This will create tensions, since it is for the Ministry for Industry, Energy and Tourism to issue said certificate. It might well happen that within discretionary powers of the Ministry there are delays in issuing the previously mentioned certificates and, as a result, the facility does not meet the 1 January deadline and, consequently, does not become entitled to capacity payments.

21.3.5 Judicial review of regulations introducing capacity payments In Spain, subjects enjoying locus standi may apply for judicial review of either administrative acts or administrative norms. When they are adopted by central governmental institutions, the competent court will be either the Audiencia Nacional or the Tribunal Supremo, the Spanish Supreme Court. Decisions of the Audiencia Nacional, in normal circumstances, can be appealed in front of the Tribunal Supremo. Traditionally, judicial review courts have been reluctant to quash decisions based on discretionary powers of the government, but legal progress is leading to an increasing readiness of courts to examine whether the government has abused its discretionary powers and acted in an arbitrary manner. Initially, according to the MO 1998,37 power plants were required to operate for a minimum of 480 hours per year in order to be eligible for capacity payments. This was reduced to fifty hours by the Royal Decree 1454/2005.38 This Royal Decree was appealed in front of the Tribunal Supremo. It was argued that, first of all, the measure in question breaches the principle of legal certainty. Lowering of the number of hours means that each production unit will receive less money, since there would be more units entitled to it, whereas the pool of money available for capacity payments remains the same. Second of all, the applicant argued that the measure had a retroactive effect. This appeal was dismissed in 2007 by the Tribunal Supremo. In view of the Supreme Court, the measure has no retroactive effect, as it will apply from the moment the Royal Decree 1454/2005 is published in the Official Bulletin. The Supreme Court found also that the principle of legal certainty has not been violated, as generators do not have a right to an unchanged system of capacity payments. According to the Supreme Court, the government can exercise its regulatory powers and modify the criteria and conditions of capacity payments.39 In the light of the initial wording of Article 16 of the ESA 1997, which did not differentiate between technologies eligible for capacity payments, two electricity companies appealed RD 1634/2006,40 because it stated that nuclear power plants will not receive capacity payments. In 2009 the Supreme Court found that the appellants were 37 38

MO 1998 (n 25). Royal Decree No 1454/2005 of 2 December 2005, Spanish Official Bulletin No 306 of 23 December

2005. 39 40

2006.

Tribunal Supremo, Decision of 10 October 2007 in Appeal No 14/2006. Royal Decree No 1634/2006 of 29 December 2006, Spanish Official Bulletin No 310 of 30 December

21.3 Capacity mechanism

361

right and declared the appealed provision of RD 1634/2006 to be null and void. It stated that an administrative norm passed by government (RD 1634/2006) could not run counter to a parliamentary act (ESA 1997), since this act did not give any ground to discriminate among technologies.41 An electricity company and the UNESA (the association of Spanish utilities) submitted two appeals against the MO 2794/2007,42 arguing that its provisions are discriminatory. According to MO 2794/2007, only power plants with an installed capacity of 50 MW or higher, which started operation after 1 January 1998 and are not older than ten years are eligible for capacity payments. The appeal was denied by the Supreme Court in 2011. The Supreme Court found that there is no discrimination. In the Supreme Court’s view, the appellants’ grounds for appeal are not sufficiently elaborated. In other words, although the applicant argued that some principles had been violated, it did not go further to explain why and how. The appellants stated that the MO 2794/2007 created a competitive advantage of some producers over others, which could be considered state aid. Further, the appellants found that the MO 2794/ 2007 is contrary to the 2003 Electricity Directive43 due to its impact on price formation. The appellants argued that the 2003 Electricity Directive only allows for a differentiated remuneration if security of supply is at risk.44 The appellants also challenged the MO 2794/2007 under the Spanish Competition Act,45 as different payments to different types of installations may distort competition in the domestic market. Finally, the appellants claimed that the difference in payments would be contrary to the ESA 1997 according to which the remuneration of the power sector activities must meet criteria of objectivity, transparency, and non-discrimination.46 The Supreme Court found that the applicant did not make any effort in further developing its arguments. The fact that the MO 2794/2007 sets a starting date (1 January 1998) to determine which installation is entitled to capacity payments does not, in view of the Court, relate to any of the appellants’ arguments. The Court was of the opinion that provisions of the MO 2794/ 2007 remained within the government’s margin of discretion and appellants had failed to provide sufficient evidence to support their appeal.47 Another appeal was submitted against MO 2794/2007. The appeal was based on an alleged breach of the principle of legitimate expectations. The Supreme Court found in 2011 that the principle had not been violated, since changes in the system of capacity payments were due to the need to implement in Spain the 2003 Electricity Directive. The principle of legitimate expectations does not guarantee that the system of capacity payments will remain unchanged, but allows the government and the regulatory authorities to modify it within the scope of their discretion and general interest.48

41

Tribunal Supremo, Decisions of 28 January 2009 in Appeals 42/2007 and 37/2007. 43 MO 2794/2007 (n 30). 2003 Electricity Directive (n 29). 44 2003 Electricity Directive (n 29) Art 3(2). 45 Act No 15 of 3 July 2007 on the Protection of Competition, Spanish Official Bulletin No 159 of 4 July 2007. 46 ESA 1997 (n 5) Art 15. 47 Tribunal Supremo, Decision of 8 March 2011 in Appeal No 90/2009 and Decision of 18 March 2011 in Appeal No 623/2009. 48 Tribunal Supremo, Decision of 22 March 2011 in Appeal No 87/2009. 42

362

Spain

UNESA also challenged MO 3860/2007, which amended MO 2794/2007. The Supreme Court confirmed that the principle of legitimate expectations should not impede the government and regulatory authorities from changing the rules whenever the general interest so requires.49 MO 3127/2011, modifying MO 2794/2007 with regard to capacity payments, was also subject to an appeal. The appellant argued that the provisions of MO 3127/2011 discriminated against hydroelectric power plants in favour of thermal plants, since payments for the former were remarkable lower. The Supreme Court found that the distinction was based on objective criteria, which had been clearly explained by the Spanish government. In the Supreme Court’s view, the review of the index values to be applied in the annual remuneration of the availability service is justified by the changing needs of the capacity payments given the impact of the economic crisis on electricity demand and thus operation of certain power plants and facilities. The constitutional principle of equality (prohibition of discrimination) is not violated, since the different treatment among electricity technologies is justified.50 MO 3127/2011 was challenged by an electricity company owning nuclear power plants, since it excluded nuclear plants from capacity payments. In this case, the appellant’s main argument was that the Supreme Court had decided in 2007 that nuclear plants should not be excluded from this capacity mechanism.51 However, the Supreme Court dismissed this appeal, arguing that MO 3127/2011 develops a new system of capacity payments, quite different from the original system set up by the ESA 1997. In this new system, capacity payments are intended to promote mediumand short-term availability of generation capacity, rather than long-term availability. Given that nuclear power plants are not flexible enough to respond to medium- and short-term needs of the electricity system, their exclusion from capacity payments is justified.52

21.3.6 Capacity payments in the 2013 Electricity Sector Act In accordance with the ESA 2013, capacity payments are subject to a new regulatory framework. The Ministry is not under an obligation to introduce capacity payments, since the ESA 2013 leaves the government discretion as to whether and when to introduce a new system of capacity payments.53 The ESA 2013 introduces a distinction between tariffs and charges and explains in its preamble that this difference in terminology is in line with the wording of the 2009 Electricity Directive.54 However, one can observe that the 2009 Electricity Directive does not provide a clear distinction between these two terms. Rather, it seems that they are used interchangeably. The ESA 2013 provides a much clearer distinction in that tariffs refer to payments to cover transmission and distribution costs and are in line 49 50 51 52 53 54

Tribunal Supremo, Decision of 7 March 2011 in Appeal No 92/2009. Tribunal Supremo, Decision of 18 July 2013 in Appeal No 87/2011. Tribunal Supremo (n 41). Tribunal Supremo, Decision of 8 November 2013 in Appeal No 44/2012. ESA 2013 (n 8) Art 13 (d). ESA 2013 (n 8) Art 13 and section II of the preamble; 2009 Electricity Directive (n 21).

21.4 European dimension

363

with the provisions of the 2009 Electricity Directive (access tariffs). To the contrary, charges are a novel concept under the ESA 2013, and means other costs of the electricity system. This includes, among others, remuneration of generation from RES, high efficiency cogeneration or residues, remuneration of production in the extra-peninsular and insular electricity systems, remuneration associated with capacity mechanisms and with annual payments to reduce effects of the electricity tariff deficit, including interests. It will be for the government to decide how charges are going to be paid for, but is seems clear that the new distinction between tariffs and charges is intended to allow the government, for example, to stop financing renewable energies via electricity prices paid by consumers, and start financing them from the state budget. The Ministry for Industry, Energy and Tourism has so far not exercised its discretion to introduce a new capacity mechanism based on the ESA 2013. Thus, the Ministerial Order IET/107 of 31 January 2014 (MO 107/2014) establishes that capacity payments for the medium-term availability service foreseen in MO 2794/2007 (as amended by MO 3127/2011) will continue to apply in 2014. As already explained, payments promoting long-term investment for new installations were abolished in 2013.

21.4 European dimension Spanish capacity payments have not raised any questions of compatibility with EU law so far, considering that these payments are available to all electricity producers operating within the Spanish electricity system regardless of their nationality. In fact, when CNMC launched a public consultation on capacity payments in 2012, none of the questions raised therein related to EU law.55 Spanish electricity generators differ in their views on capacity payments. Generally, operators of conventional generation units support capacity mechanisms, which provide them with an additional stream of revenues to recover their investment costs. According to them, ‘capacity mechanisms complement energy-only market in order to preserve efficiency in terms of generation adequacy and security of supply.’ In contrast, generators with a predominantly renewable energy portfolio are not in favour of keeping the system of capacity payments. For them, ‘capacity mechanisms should constitute an intervention of last resort, as they have to be very carefully designed not to cause significant distortions in the functioning of the internal market.’ At the same time, conventional generators are in favour of a national approach to generation adequacy, whereas operators of RES generation take a broader, European perspective.56 Some capacity mechanisms may fall within the scope of EU state aid rules. The 1997–2007 system of capacity payments was linked to the compensation of stranded costs, and the Commission has so far considered such compensation schemes

55

CNMC (the then CNE), Consulta pública sobre el mecanismo de pagos por capacidad, 24 May 2012. Quotations from the responses to the Commission’s consultation on generation adequacy (section 1.4.5) of, respectively, UNESA (the Spanish association of electricity producers) and ACCIONA (the biggest Spanish utility in the field of electricity production from RES). Responses are available at the Commission’s website at http://ec.europa.eu/energy/en/consultations/consultation-generation-adequacy-capacity-mechanisms-andinternal-market-electricity, accessed 1 February 2015. 56

364

Spain

compatible with EU law.57 Similar compensation schemes established in Denmark and Austria were also cleared by the Commission. This is explained in the following terms: ‘If the capacity payment is designed as a fixed (annual) payment to all capacity, the short term effect is merely to increase revenues for existing capacity (eg the Spanish scheme to contribute to the recovery of stranded costs).’58 The reductions in capacity payments and their partial abolishment introduced by RDL 9/2013 further minimizes the risk of breaching EU state aid rules. However, capacity payments for the availability service should be assessed under EU state aid law. Nowadays, certain payments paid out by the state to prevent closure of some production units in times of economic crisis and lower electricity demand can be considered compatible aid. This is because those electricity production from these units can be considered SGEI in that they constitute back-up power plants to intermittent RES generation. In broader terms, capacity payments to prevent the closure of existing production units should be considered a compensation for PSO (ensuring security of supply) and as such, compatible with EU law. Moreover, if capacity payments provide back-up generation for RES and in this way promote further RES integration (as in the case of the Spanish capacity payments), they even promote EU energy policy goals, and should not be considered in breach of EU law. Following substantial changes introduced to the Spanish system of capacity payments, the current regulatory framework seems to address the key design issues put forward by Hancher, Sadowska, and Willems: payments are not paid out unconditionally, they do not discriminate between domestic and foreign generators, and distortion of electricity spot prices is limited.59

21.5 Conclusion Capacity payments have played an important role within the Spanish electricity regulatory framework since the 1997 market liberalization. They need to be considered in the context of other provisions aiming to smoothen the transition from a centrally planned economy to a market system, in particular the provisions on the compensation of stranded costs from 1997 to 2006. Capacity payments should be understood as an instrument to guarantee adequate investments in generation and transmission. Since 2007, capacity payments have experienced a radical change. In contrast to the initial regulation, the ESA 2013 allows the government to decide whether to support generators with capacity payments or not. Substantial reductions in capacity payments over the recent years (and their partial abolishment) have been driven by practical considerations, such as the bad financial situation of the Spanish electricity sector, rather than EU law compliance. 57 European Commission, Decision SG D/290553 of 25 July 2001 adopted in Case SA NN 49/99 Spanish stranded costs. In this decision the Commission authorized Spain to grant compensation for stranded costs until 2008 to the companies which were asked to pre-finance the 2005 deficit. 58 THEMA consulting group (n 28) p 49. 59 Leigh Hancher, Małgorzata Sadowska, and Bert Willems, Generation adequacy, capacity mechanisms and the internal market in electricity, reply to the 2012 consultation on generation adequacy, available at the Commission’s website (n 56).

22 United Kingdom Peter Willis

22.1 Introduction The first auctions in the new UK capacity mechanism took place in December 2014, for delivery in 2018–2019. They followed an exhaustive four-year process of analysis and consultation, which has produced what must be one of the most carefully designed and thoroughly scrutinized capacity mechanisms in the world. This chapter outlines the evolution of the UK capacity mechanism (more precisely referred to as the ‘GB capacity mechanism’)1 including the reasons that made it necessary, the design options considered, the key features of the mechanism finally adopted, and the important EU dimension. It should be noted that the documents published on this issue by the government and key stakeholders run to many thousands of pages, and the following chapter can provide only a brief overview of the topic.

22.2 Setting the scene 22.2.1 Market characteristics The GB market was liberalized in the 1990s. Generators, suppliers, traders, and large customers currently buy and sell electricity on an energy-only wholesale market, by means of bilateral contracts, as well as over the counter and on spot markets.2 The GB TSO, National Grid, operates the onshore and offshore transmission system, and owns the transmission system in England and Wales. The transmission system in Scotland is owned by SSE and Scottish Power, two of the six major integrated energy groups. Electricity is traded in thirty-minute settlement periods. At gate closure, one hour before the start of the settlement period, National Grid becomes responsible for balancing supply and demand through the balancing mechanism. In 2012, the total electricity generated in the UK was 364 TWh. Of this, 39 per cent was generated by coal, 28 per cent by gas, 19 per cent by nuclear and 11 per cent by renewables. Net imports were 12 TWh, or 3 per cent of the total electricity supply.3 1 Great Britain (GB) means England, Wales, and Scotland. Northern Ireland, which is also part of the United Kingdom, forms part of the Single Electricity Market with the Republic of Ireland and falls outside GB Electricity Market Reform proposals. 2 These ‘British Electricity Trading and Transmission Arrangements’ were introduced (as the ‘New Electricity Trading Arrangements’) in England & Wales in 2001, and extended to cover Scotland in 2005. They replaced the former pool mechanism which included a capacity element. 3 Digest of United Kingdom energy statistics (DUKES), Chapter 5: Electricity, available at https://www. gov.uk/government/publications/electricity-chapter-5-digest-of-united-kingdom-energy-statistics-dukes, accessed 1 February 2015.

366

United Kingdom

22.2.2 Regulatory framework The GB electricity market is governed by a complex body of legislation consisting primarily of the Electricity Act 1989 as amended and supplemented by the Utilities Act 2000, the Energy Act 2004, the Energy Act 2008, the Energy Act 2010, the Energy Act 2011, and the Energy Act 2013,4 as well as a large number of pieces of secondary legislation. Most generators, suppliers, and network owners/operators are licensed, and their licences impose various obligations, including compliance with applicable industry codes. OFGEM, the energy regulator, regulates energy companies through the licence regime. It also enforces the competition rules in the energy sector.

22.2.3 Generation adequacy 22.2.3.1 Background The government (the Department of Energy & Climate Change, DECC) argues that the GB market structure has been effective in developing a competitive energy market over the past twenty years. However, early in 2010, OFGEM reported on the results of its ‘Project Discovery’, a year-long review of GB medium-term security of supply prospects.5 It concluded that urgent and significant action was needed in order to guarantee GB security of supply. It proposed a series of possible policy reforms. Those options included the introduction of tenders for capacity for some or all generation as well as demand side response (DSR). The government built on this foundation in its December 2010 Consultation Document ‘Electricity Market Reform’.6 The 2010 Consultation Document highlighted the need to replace a quarter of existing generation capacity by 2020, against a background of increasing demand for electricity and a requirement to decarbonize generation. In the 2010 Consultation Document, the government proposed an Electricity Market Reform package consisting of four elements: (a) carbon price support, (b) feed-in tariffs, (c) capacity payments, and (d) an emissions performance standard. The four elements of EMR are closely interlinked, but only the third, the development of a capacity mechanism, is discussed here.

22.2.3.2 Reasons for the gap In a series of consultations and policy documents,7 the government identified8 a number of market failures which had led to the emergence of a capacity gap. It 4

All acts are available at http://www.legislation.gov.uk, accessed 1 February 2015. OFGEM, Project Discovery—options for delivering secure and sustainable energy supplies, Consultation document 16/10, 3 February 2010. 6 DECC, Electricity Market Reform, Consultation Document, December 2010 (2010 Consultation Document). 7 A notable feature of the development of the UK capacity mechanism is the very significant consultation process, over a period of more than three years, involving several rounds of consultation and proposal, and many thousands of pages of documents published by the government. 8 See, for example, DECC, 2010 Consultation Document (n 6) paras 35–6. 5

22.2 Setting the scene

367

highlighted three of these as particularly important. The first was the status of reliability as a public good. Consumers are currently unable to choose the level of reliability (protection from disconnection) that they require, and to pay accordingly. Significant wholesale price increases may therefore be needed in order to secure all consumers against disconnection. Secondly, barriers to entry in the wholesale market, including a lack of liquidity, have limited the entry of new capacity, as has, thirdly, the fact that wholesale electricity prices may send inadequate investment signals to ensure the development of the necessary flexible plant. The government identified the third of these points as the most significant. It argued that this was in turn due to what it described as the ‘missing money’.9 This was a result of a number of factors. In particular, peak wholesale prices did not necessarily cover flexible generators’ long-run marginal costs, and developers were therefore reluctant to invest in new capacity. One reason for this was that the cost of balancing actions taken by the TSO was not fully reflected in the ‘cash-out’ imbalance price.10 However, even if cash-out prices had reflected costs fully and provided the correct incentives, investors were concerned that high prices might lead to price caps. More widely, there was general uncertainty about GB energy policy, as a result of the significant number of reforms in recent years. Investors were accustomed to dealing with market forces, but less comfortable when faced with prices that were heavily influenced by governments. This led to a ‘wait and see’ approach. At the same time, investment cycles tended to lead to peaks and troughs of capacity, even where the capacity over a longer period was adequate.

22.2.3.3 Scale of the gap The government put significant effort into assessing the size of the capacity gap. It assessed the gap by reference to the peak de-rated capacity margin and the expected energy unserved (EEU). The peak de-rated capacity margin is measured by adjusting total capacity to take account of availability, specific to each type of generation technology. It reflects the expected proportion of a source of electricity which is likely to be technically available to generate. EEU is an assessment of the likelihood of a disruption and its likely size. The government noted that UK EEU arising from distribution system faults was approximately 12 GWh per year, compared with total

9 This is quantified in some detail in DECC, Electricity Market Reform—Capacity Market Impact Assessment, IA No DECC0103, 27 November 2012 (November 2012 IA). 10 Electricity generators and suppliers are responsible for balancing their own positions through bilateral contracting and trading, leaving National Grid to resolve the remaining imbalance. The ‘cash-out’ price is the price paid by a party that is in imbalance. This price not only directly affects the price that consumers pay for their energy, but also provides the incentives on the electricity market participants to invest in secure supplies. The 2010 consultation document noted that the highest cash-out price to date was 938 £/ MWh, whereas a price of some 10,000 £/MWh might be required in order to reflect costs fully and provide the correct incentives. The then cash-out price reflected the price of the last 500 MWh, rather than that of the marginal plant. OFGEM has since proposed reforms (see section 22.2.3.4) that will result in the cash-out price more closely reflecting the last (and highest-priced) balancing actions, raising cash-out prices to as much as 6,000 £/MWh.

368

United Kingdom

UK electricity supply of around 364 TWh. The UK EEU from generation-related faults was near zero.11 In its 2010 Consultation Document, DECC estimated that without a capacity mechanism, spare capacity would fall to around 5–11 per cent between 2020 and 2030, leading to an EEU resulting from insufficient generation of around 0.5–7 GWh.12 The government noted that it was necessary to strike a balance between optimizing security of supply and the cost borne by consumers. It considered that an economically optimum peak de-rated capacity margin might be around 8–12 per cent, resulting in an EEU of around 0.5–4 GWh per year. The cost was based on the average VOLL which is the theoretical value of preventing supply disruption. It is the price at which an average consumer would prefer to be cut off than continue to pay for electricity.13 At that time, the government assumed a VOLL of 10,000 £/MWh, although it subsequently revised that figure to 17,000 £/MWh.14 By the time it issued its July 2011 White Paper, DECC had also revised its estimates of the capacity margin.15 It estimated that even without the market failures identified earlier, the capacity margin might drop to 2–8 per cent between 2020 and 2030, resulting in an EEU of 0.5–25 GWh. A 3 per cent capacity margin could result in an EEU of 20 GWh, with estimated costs to the economy of £200–600 million.16 If prices were not allowed to rise to VOLL, or investors believed that they would not be, the margin might fall still further. DECC revised its figures again the following year. In its November 2012 Design and Implementation Update, the government’s ‘base case’ expected that the margin would fall to 1 to 15 per cent (ie a capacity margin potentially falling to less than zero) between 2020 and 2030.17 Its ‘stress test’ (assuming higher demand, some ‘missing money’ and delays to new nuclear build and offshore wind capacity in the early 2020s) resulted in a capacity margin of 4 to 8 per cent between 2020 and 2030. It illustrated the capacity margin as shown in Figure 22.1. The government commented in its November 2012 Impact Assessment18 that: The key points to take away from looking at the range of modelling we have undertaken is that (a) there remains a credible risk of a capacity problem in the medium-term; however (b) the further into the future we try to assess future levels of capacity, the less certainty we have about the outcome.

It then revised its assessment of the margin further in its October 2013 Impact Assessment,19 based on OFGEM’s Electricity Capacity Assessment in June 2013,20

11

DECC, 2010 Consultation Document (n 6) p 30. DECC, 2010 Consultation Document (n 6) p 30. 13 14 VOLL (Value of Lost Load) is defined in chapter 1 at n 7. See section 22.3.3.1 below. 15 DECC, Planning our electric future: a White paper for secure, affordable and low-carbon electricity, July 2011 (White Paper). 16 DECC, White Paper (n 15) para 3.2.13. 17 DECC, Annex C—Capacity Market: Design and Implementation Update, November 2012. 18 DECC, November 2012 IA (n 9) para 3.35. 19 DECC, Electricity Market Reform—Capacity Market Impact Assessment, October 2013 (October 2013 IA). 20 OFGEM, Electricity Capacity Assessment Report 2013, 27 June 2013. 12

22.2 Setting the scene

369

25% 20% 15% 10% 5%

20 12 20 13 20 14 20 15 20 16 20 17 20 18 20 19 20 20 20 21 20 22 20 23 20 24 20 25 20 26 20 27 20 28 20 29 20 30

0% –5% –10% DECC base case: central demand, some interconnector imports DECC stress test: higher demand, some interconnector imports, delays to low carbon, £500/MWh price cap Ofgem base case: higher demand, some interconnector exports

Figure 22.1 OFGEM and DECC estimates of de-rated capacity margins Source: DECC, Annex C—Capacity Market: Design and Implementation Update, November 2012. This and the other figures in this chapter contain public sector information licensed under the Open Government Licence v3.0: http://www. nationalarchives.gov.uk/doc/open-government-licence/version/3/

and OFGEM’s and DECC’s joint Statutory Security of Supply Report 2013.21 It noted that the risks to security of supply were likely to increase faster than anticipated in October 2012. The October 2013 Impact Assessment reported on a detailed modelling process based on the government’s own Dynamic Dispatch Model, and covering a range of assumptions and sensitivities, including various interconnection scenarios. Most recently, the government’s assessment of the margin in its September 2014 Impact Assessment,22 based on OFGEM’s Electricity Capacity Assessment in June 2014,23 highlighted an expected further reduction in margins, but at the same time the introduction of new balancing services that would secure additional capacity.24

22.2.3.4 Minimizing the need for a capacity mechanism—improving the current market Recognizing that a capacity mechanism is an artificial market, the government proposed a number of measures to reduce the capacity gap and thereby minimize the need 21 DECC and OFGEM, Statutory Security of Supply Report: 2013, Ref ISBN 9780102986662, 13D/266, HC 675 2013–14, 31 October 2013. 22 DECC, Electricity Market Reform—Capacity Market Impact Assessment, September 2014 (September 2014 IA). 23 OFGEM, Electricity Capacity Assessment Report 2014, 30 June 2014. 24 National Grid is introducing a Supplemental Balancing Reserve and a Demand Side Balancing Reserve, approved by OFGEM in December 2013. See OFGEM’s decision of 19 December 2013, empowering National Grid to introduce new balancing services including payments to firms for creating reserves on the demand side, available at https://www.ofgem.gov.uk/publications-and-updates/national-grid%E2%80% 99s-proposed-new-balancing-services-decision-letter, accessed 1 February 2015.

370

United Kingdom

for intervention.25 These measures include reforms to the balancing arrangements26 and actions to improve diversity and the demand side. In particular, the government proposed reforms to the calculation of cash-out payments, so that prices are permitted to rise to as much as 6,000 £/MWh, more accurately reflecting the costs of balancing the system and sending stronger investment signals, improving the procurement of balancing services, taking action to manage intermittent renewables and further actions to improve liquidity.27 It also proposed measures to improve diversity and the demand side. These include promoting demand side response, interconnection, storage, and energy efficiency.

22.3 Capacity mechanism 22.3.1 Background 22.3.1.1 Assessment of the options The design of the GB capacity mechanism has been an evolutionary process, with the government initially preferring one possible model and finally deciding on another following consultation. The government set out the options in its December 2010 Consultation Document.28 It noted that the capacity margin is in effect currently determined by market forces. For the reasons set out earlier, this had failed to secure adequate capacity. An alternative was therefore for the government to determine the required capacity margin, and then to incentivize the provision of capacity to ensure that the required margin is met. Instead of flexible generators obtaining their revenue exclusively from the supply of electricity at peak prices, as was the case under the former ‘energy-only’ market, under a capacity mechanism they would obtain part of their revenue from the provision of capacity. This would supply the ‘missing money’, ensuring that by means of a combination of energy and capacity payments, they would recover their long-term marginal costs. In the December 2010 Consultation Document, the government highlighted the numerous ways in which a capacity mechanism could be designed in order to achieve this objective by, where appropriate, citing examples from other jurisdictions in the EU and USA.29 Among the different capacity mechanism types, which have been already discussed in section 1.2.3 above, DECC also considers a tender for targeted resource (TTR), under which capacity payments are made only to those resources needed to make up any shortfall in the market. The level of payment is set through a competitive tendering process. Conditions imposed on the operation of the resource limit the market distortion. An example is the Swedish peak-load reserve mechanism. 25

DECC, 2010 Consultation Document (n 6) pp 79–84. DECC, 2010 Consultation Document (n 6) pp 79–84. 27 OFGEM launched a Significant Code Review on this issue in March 2012, and issued a policy decision in May 2014. See OFGEM, Electricity Balancing Significant Code Review—Final Policy Decision, 15 May 2014, available at https://www.ofgem.gov.uk/ofgem-publications/87782/electricitybalancingsignificantcodereviewfinalpolicydecision.pdf accessed 1 February 2015. 28 DECC, 2010 Consultation Document (n 6). 29 DECC, 2010 Consultation Document (n 6) pp 86–99. 26

22.3 Capacity mechanism

371

According to DECC, a strategic reserve (discussed in section 1.2.3.1) is a subset of this model, in which the resource is used only after all other resource has been exhausted, as opposed, for example, to the extension of the GB Short-Term Operating Reserve (STOR) mechanism.30 Figure 1.2 in section 1.2.3 illustrates the relationship between these options. In the December 2010 Consultation Document, the government expressed an initial preference for a targeted capacity mechanism.31 It explained that a market-wide mechanism would mean a fundamental change to the GB electricity market. Given the scale of change, there would be a risk of uncertainty and therefore delayed investment. A targeted mechanism would involve a smaller intervention in the market (with an estimated 5 GW of capacity required in order to maintain a margin of 10 per cent) and would therefore create a lower risk of market distortion. A targeted mechanism would build on existing GB experience with the STOR and the Swedish experience with the peak-load reserve. A targeted mechanism would also be more flexible and more capable of responding to changing market conditions. The government acknowledged that a targeted capacity mechanism bore some risks of market distortion, through two effects. First, capacity payments might not feed accurately into the cash-out price and therefore might not secure an adequate return. Secondly, a targeted mechanism would be susceptible to the ‘slippery slope’ effect, by which operating exclusively in the capacity mechanism, outside the market, might become more attractive than remaining in the market, leading to a lack of investment in the market and a requirement to procure ever greater capacity through the capacity mechanism. Nevertheless, provided that these risks of market distortion could be addressed, the government considered that a targeted mechanism was more attractive.

22.3.1.2 Assessment of the strategic reserve and capacity market options Respondents to the consultation expressed concerns about the targeted mechanism. The ‘slippery slope’ effect was a particular concern for some of them. In its July 2011 White Paper,32 DECC therefore proposed refinements to the targeted mechanism, involving the creation of a strategic reserve as described earlier. However, it also explored an alternative market-wide capacity market model.33 It assessed the two options in more detail in a fifty-page annex to the White Paper.34 It described the operation of the capacity market option as follows:35

30

STOR is a service for the provision of additional active power from generation and/or demand reduction. It provides National Grid with reserve power in the form of either generation or demand reduction so that it can deal with actual demand being greater than forecast demand and/or plant unavailability. Where it is economic to do so, National Grid will procure part of this requirement ahead of time through STOR. See http://www2.nationalgrid.com/uk/services/balancing-services/reserve-services/ short-term-operating-reserve/, accessed 1 February 2015. 31 DECC, 2010 Consultation Document (n 6) paras 66–79. See also Figure 1.2 in section 1.2.3. 32 DECC, White Paper (n 15). 33 DECC described the two proposals at that stage in its White Paper (n 15) pp 73–6. 34 DECC, Annex C—Consultation on possible models for a Capacity Mechanism, 12 July 2011 (Annex C). 35 DECC, Annex C (n 34) p 180.

372

United Kingdom

The required volume of reliable capacity would be determined centrally based on forecasts of the peak demand some years ahead. That total amount of demand for capacity would be purchased from any provider willing to supply it, subject to its ability to meet the necessary criteria. Providers of capacity could include existing generators, companies that are planning to build a new power plant, and companies offering other forms of capacity such as DSR or storage. In effect, providers of capacity in a capacity market substitute uncertain returns in the electricity market for longterm certainty from the capacity market. Consumers benefit from certainty of supply and increased price stability.

DECC identified two issues that a capacity market must address: how to decide how much capacity can be offered by a particular plant, and what penalties to impose if the promised capacity is not available. This can be addressed either by an administrative process or, by supporting the mechanism with reliability options, ie market-based incentives such as financial call options (see section 1.2.3.4).

22.3.1.3 Selection of the capacity market as preferred option By December 2011, the government had settled on a capacity market as the preferred option. In its Technical Update to the White Paper, it noted that of the seventy-four responses to the White Paper consultation, 35 per cent preferred a market-wide mechanism and 25 per cent preferred a strategic reserve. Significantly, five of the six large integrated energy groups and four of the eight independent generators that responded preferred a market-wide capacity mechanism to the strategic reserve option. The government explained that the strategic reserve option would require wholesale electricity prices to spike to very high levels. This would mean keeping potentially efficient plants outside the market (because the plants would be kept as a last resort). It would also lead to concerns that the government or regulator would intervene to reduce prices. Finally, the government confirmed that notwithstanding the fact that the strategic reserve option had been proposed in order to reduce the risk, it shared the concerns of many of the respondents to the consultation that the strategic reserve option might lead to a slippery slope in which more and more capacity was pulled into the reserve.36 However, the government recognized that some respondents found it difficult to debate the detail of the capacity market without high-level decisions having been taken. It therefore explained that it would work on the detailed design of the mechanism in the next phase, with other stakeholders.

22.3.1.4 Refinement of the capacity market design The government set out the high-level design of the capacity market in November 2012, and then Detailed Design Proposals in June 2013,37 together with a Capacity 36

DECC, Planning our electric future: technical update, December 2011. DECC, Electricity Market Reform: Capacity Market—Detailed Design Proposals (2013) (Detailed Design Proposals). 37

22.3 Capacity mechanism

373

Market Strawman.38 It then issued a Consultation on Proposals for Implementation in October 2013,39 and an ERM Delivery Plan in December 2013,40 each accompanied by numerous supporting documents.

22.3.2 Legislative framework Details of the GB capacity market, as it evolved in the period of more than three years since it was first proposed, were set out in the series of consultation and policy papers, and the delivery plan referred to earlier. In legal terms, the capacity market is addressed in chapter 3 of the Energy Act 2013.41 Chapter 3 entered into force on 18 December 2013. It is enabling legislation, allowing the government to adopt detailed rules through secondary legislation. For example, section 27(1) of the Energy Act 2013 provides that ‘[t]he Secretary of State may by regulations make provision for the purpose of providing capacity to meet the demands of consumers for the supply of electricity in Great Britain.’ In October 2013, DECC consulted on draft secondary legislation that set out the detailed rules for the capacity market,42 which the government adopted as the Electricity Capacity Regulations 2014,43 the Capacity Market Rules 2014,44 and the Electricity Capacity (Supplier Payment) Regulations 2014,45 to provide for payments by suppliers financing the capacity market.

22.3.3 Structure of the capacity market The capacity market will involve six phases: determination of the capacity required, eligibility and pre-qualification, auction, secondary market, delivery, and payment. These are usefully illustrated in the October 2013 Consultation document (see Figure 22.2).46

38 DECC, Capacity Market Strawman v11, June 2013. Properly, a strawman is a weak argument that is set up in order to be rebutted. Instead, the DECC strawman describes itself as a ‘working document, intended to provide a technical description of how the proposed GB Capacity Market might work in practice’. 39 See DECC, Electricity Market Reform: Consultation on Proposals for Implementation (document ref ISBN 9780101870627, 10 October 2013). The consultation document, draft legislative proposals as well as results of this consultation are available at the consultation website https://www.gov.uk/government/ consultations/proposals-for-implementation-of-electricity-market-reform, accessed 1 February 2015. 40 DECC, Electricity Market Reform Delivery Plan, Policy paper, 19 December 2013 (ERM Delivery Plan). 41 Energy Act 2013 (n 4). 42 All (draft) legislative proposals are available at DECC’s consultation website (n 39). 43 Electricity Capacity Regulations 2014, SI 2014/2043, 31 July 2014 (n 4). These replaced proposed separate Electricity Capacity Regulations and Electricity Capacity (Payment) Regulations. 44 Capacity Market Rules 2014, as amended by the Capacity Market (Amendment No 1) Rules 2014 and the Capacity Market (Amendment No 2) Rules 2014, available at https://www.gov.uk/government/ publications/capacity-market-rules, accessed 1 February 2015. 45 The Electricity Capacity (Supplier Payment etc.) Regulations 2014, SI 2014/3354, 17 December 2014 (n 4). 46 DECC, EMR: Consultation on Proposals (n 39) para 381, Figure 4.3.

374 Amount to auction Enduring reliability standard established by Govt. System Operator develops scenarios of peak demand, and advises on the amount of capacity needed to meet the reliability standard.

United Kingdom Eligibility and prequalification Demand side response and storage eligible as well as generation. Mandatory for all licensed generators to go through prequalification process or submit an opt-out notification.

Auction Central auction held to set the price for capacity and determine which providers are issued with capacity agreements

Trading Capacity providers may adjust their position in private markets.

Delivery

Payment

Providers of capacity commit to be available when needed or face penalties in the delivery year.

Costs of capacity shared between suppliers, in proportion to their share of peak demand.

Capacity Market does not replace electricity market.

Figure 22.2 Stages of capacity market auction Source: DECC, EMR: Consultation on Proposals (n 39) para 381, Figure 4.3.

22.3.3.1 Stage 1—determination of the capacity required The government establishes a reliability standard. The purpose of the reliability standard is to set the level of adequacy that the capacity market is to provide. It is intended to reflect a trade-off between the level of reliability required in a system and the cost of ensuring that reliability. The reliability standard takes the form of a LOLE, ie a statistical expression of the level of reliability expected in the GB market.47 In its EMR draft Delivery Plan, published in July 2013, the government consulted on a proposed reliability standard of three hours per year, equating to a system security level of 99.97 per cent.48 It took this figure forward in its December 2013 ERM Delivery Plan.49 The LOLE is calculated by reference to the ratio of the cost of building a new marginal peaking plant (Cost of New Entry—CONE) to the value that consumers place on avoiding disruption (VOLL).50 The government used the cost of a new CCGT plant to calculate the CONE.51 Establishing the VOLL was a complex process. Although earlier DECC publications indicated a figure of 10,000 £/MWh, the July 2013 Consultation on the draft Delivery 47 LOLE stands for the loss-of-load expectation and is one of the possible indices to measure the level of supply reliability. See section 5.5.1 for an explanation. 48 DECC, Consultation on the draft Electricity Market Reform Delivery Plan, URN 13D/139, July 2013. 49 DECC, ERM Delivery Plan (n 40). 50 VOLL (Value of Lost Load) is defined in chapter 1 at n 7. 51 The government had originally based the CONE on the costs of a new OCGT plant. However, it concluded that it was unlikely that new large OCGT plants could be built between the date of the first auction and the first delivery year, so it based Net CONE for the first auction on the estimated level at which new CCGT could bid into the capacity market. It reflects the cost of CCGT, less expected electricity market and ancillary service revenue. This caused net CONE to increase from 29 £/kW to 49 £/kW.

22.3 Capacity mechanism

375

Price £/kW year

Cap

Net CONE

Capacity GW Minimum T–1.5GW or T–5%

Target T

Maximum T+1.5GW or T+5%

Figure 22.3 Illustrative capacity demand curve Source: DECC, EMR: Consultation on Proposals (n 39).

Plan set out DECC’s proposed figure of 17,000 £/MWh.52 This was based on a 225-page report on the Value of Lost Load prepared for DECC and OFGEM by London Economics.53 This calculation then produced the LOLE reliability standard figure of three hours/year. This was the definitive figure adopted in the December 2013 Delivery Plan.54 In October 2014, the government published a letter to National Grid setting out the demand curve parameters, including the target capacity to procure in the first auction (48,600 MW), the auction price cap, and the net CONE of 49 £/kW.55 This was based on National Grid’s Electricity Capacity Report 2014, and on the EMR Panel of Technical Experts’ Final Report on the National Grid Report, which assessed the amount of capacity required for a delivery year in order to meet the specified reliability standard. An illustrative capacity demand curve is shown in Figure 22.3.56

22.3.3.2 Stage 2—eligibility and pre-qualification A range of types of capacity are eligible to participate in the capacity market: new and existing generation capacity, DSR (including embedded generation), and electricity storage.57 52

DECC, Consultation on the draft EMR Delivery Plan (n 48). London Economics, The Value of Lost Load (VOLL) for Electricity in Great Britain—Final Report for OFGEM and DECC, July 2013. 54 DECC, EMR Delivery Plan (n 40) paras 156–61. See also DECC, Annex C—Reliability Standard Methodology, available at the DECC’s consultation website (n 39). 55 DECC, Confirmation of demand curve parameters for the first capacity auction, Letter to National Grid of 13 October 2014. The target capacity to procure was reduced from the original figure of 50,800 MW to reflect opt-out decisions by a number of plants. 56 DECC, EMR Panel of Technical Experts Final Report on National Grid’s Electricity Capacity Report (2014). 57 DECC, Implementing Electricity Market Reform (ERM)—Finalised policy positions for implementation of EMR, June 2014. 53

376

United Kingdom

The government was particularly keen to ensure that DSR could participate in the capacity market, and therefore provided for one year ahead auctions to facilitate participation by DSR that would find it difficult to commit to providing capacity four years ahead. It also put in place transitional arrangements for DSR in advance of the first auction. There are specific provisions for DSR in the Capacity Market Rules58 to facilitate its participation. Eligibility for the transitional arrangements will be limited to certain types of participant in order to encourage new and recent entrants. A certain number of types of capacity are excluded. First, capacity receiving support through certain schemes such as the Renewables Obligation, Contracts for Difference or small-scale Feed-in Tariffs and renewable heat incentive are excluded. Secondly, plants contracted under certain long-term contracts for the STOR are also not allowed to participate in the market, because payments under these contracts already reflect their level of utilization, and capacity payments in addition to the existing contracted payments would result in double payments. Thirdly, in the 2014 auction at least, interconnected non-GB capacity and the interconnectors themselves do not participate. The government spent some time looking for a way for interconnected capacity to participate from the 2015 auction. This issue is addressed in more detail in section 22.4.2 below. Further, this refers also to capacity below a 2MW threshold, which will be able to participate only if combined with other capacity. Lastly, electricity demand reduction does not participate, in the 2014 auction at least, although it may be included at a later date. Although participation in capacity auctions is voluntary, licensed generators must either apply to pre-qualify their eligible plant or submit an opt-out notification. Other providers such as DSR providers must also pre-qualify. Detailed rules on issues such as eligibility and pre-qualification, operation of the auction, capacity agreements, delivery, trading, and monitoring, together with templates for a number of the forms and certificates required, are set out in the 151-page Capacity Market Rules 2014.59 62 GW of capacity qualified for the first auction. There is a dispute resolution process for participants considered ineligible as a result of the pre-qualification process. Rules for the dispute resolution process are set out in the Capacity Market Rules. OFGEM, which will hear any disputes, has published guidance on the subject.60

22.3.3.3 Stage 3—the auction This is held four years ahead of delivery, with a further auction one year ahead to permit participation by DSR providers, as noted earlier, and to refine the capacity levels. The first auction was held in December 2014. The auction is on a ‘pay as clear’ basis, so all participants will receive the clearing price set by the marginal bidder. The government believed that this would be more likely to incentivize suppliers to bid their true economic cost of providing capacity. The 58

59 Capacity Market Rules 2014 (n 44). Capacity Market Rules 2014 (n 44). OFGEM, Electricity Market Reform dispute resolution guidance, 28 August 2014, available at https://www.ofgem.gov.uk/ofgem-publications/89569/emrdrguidance.pdf, accessed 1 February 2015. 60

22.3 Capacity mechanism

377

auction is on a ‘descending clock’ basis, so that bids progressively fall until the auction discovers the lowest price at which capacity can be provided. In order to mitigate market power, bidders are classified as either ‘price takers’ or ‘price makers’. New entrants and DSR resources are classified as price makers, meaning that they are free to bid up to the overall auction price cap (75 £/kW for the first auction). Existing plant must be price takers, unless they can provide justification for why a higher price is required. Price takers can bid up to a specified threshold (25 £/kW, or 50 per cent of net CONE). The reason is that the government believes that most existing plant will have lower costs, but that a price taker threshold at this level will capture 80 per cent of existing plant.61 If successful at auction, an existing plant is awarded a one-year capacity agreement. Existing plants requiring major refurbishment may be eligible for a three-year capacity agreement. New plants are eligible for a longer period—the government had originally proposed up to ten years, and then increased that figure to fifteen years in order to reduce prices by reducing the need for investors to recover their capital costs over a shorter period. Eligibility for these longer term agreements is based on the level of planned capital expenditure. For the 2014 auction, the government set a threshold of 125 £/kW for three-year agreements and 250 £/kW for fifteen-year agreements. These figures were intended to ensure that plants undergoing routine cyclical maintenance will not be eligible for long-term contracts and that existing plants are eligible for longterm agreements only if they are spending as much on capital as it would cost to build an entirely new plant. There are various milestones and other safeguards to ensure that relevant work is in fact carried out.

22.3.3.4 Stage 4—secondary market Providers may physically trade their obligations from a year ahead of the start of the delivery year where there is additional pre-qualified capacity that can take their place. They will require the consent of the TSO. Other conditions also apply. The parties may be able to hedge their positions in financial markets at any time.

22.3.3.5 Stage 5—delivery Capacity agreements require providers to deliver a specified quantity of electricity in system stress periods. The obligation is ‘load-following’, so that if a stress event occurs when total demand is at 70 per cent of anticipated peak, providers will be required to deliver only 70 per cent of their contracted capacity. National Grid will issue Capacity Market Warnings at least four hours ahead of an anticipated stress event. Providers that do not deliver sufficient energy when required, following a Capacity Market Warning, face penalties. The penalty is 1/24th of the auction clearing price, capped at 200 per cent of a provider’s marketing capacity revenue and an overall annual cap of 61 See generally, DECC, September 2014 IA (n 22), for an explanation of a number of the government’s policy decisions immediately prior to the start of capacity auctions in December 2014.

378

United Kingdom

100 per cent of annual revenues. The TSO has the right to carry out spot tests where a provider has previously failed to deliver the required level of capacity.

22.3.3.6 Stage 6—payment Payments will be the responsibility of a new government-owned settlement body, which will delegate a number of its functions to the Capacity Market Settlement Agent, Elexon, which currently administers the Balancing and Settlement code. The cost of capacity will be recovered from suppliers according to their share of peak demand. The rules governing payments to providers are set out in the Capacity Market Regulations,62 while the rules relating to payments by suppliers are set out in the Supplier Payment Regulations.63

22.3.3.7 Exiting the capacity market The government expects the capacity market to be in place for at least ten years. However, in addition to identifying the market failures that justified the creation of a capacity market, the government identified three criteria that would permit a return to an energy-only market.64 First, prices currently jump from the short run marginal cost of the peaking plant to VOLL. Better developed DSR would reduce this price volatility. However, it will take some time for an active DSR to develop, requiring the ability for domestic consumers to reduce demand in response to real-time price signals, and demand will need to be sufficiently responsive. Secondly, if liquidity increases, that will address price concerns and help investors to build new plants. Thirdly, a greater degree of interconnection will reduce the investment needed in GB capacity and will also increase diversity of supply by connecting to markets with different generation mixes and helping to smooth out shortfalls in capacity resulting from intermittency.

22.3.4 Enforcement 22.3.4.1 Enforcement mechanism In addition to the penalty regime for non-delivery of contracted capacity, the capacity market provides a means of penalizing non-compliance. Regulation 67 of the Electricity Capacity Regulations provides that OFGEM may enforce a requirement imposed on a person under the Regulations or the Rules as if it were a relevant requirement for the purposes of section 25 of the Electricity Act 1989.65 This means that OFGEM can order termination of the offending act and/or impose a penalty. The auction rules include a prohibition on capacity market participants engaging in market manipulation, and there are also requirements to maintain information barriers so as to limit the dissemination of inside information. The definitions of these terms are 62 63

Electricity Capacity Regulations (n 43) available at (n 4). The Electricity Capacity (Supplier Payment etc.) Regulations 2014, SI 2014/3354, 17 December 2014

(n 4). 64

DECC, October 2013 IA (n 19) paras 4.30–37.

65

Electricity Act 1989 (n 4).

22.3 Capacity mechanism

379

those used in REMIT.66 Auction participants will be required to sign a Certificate of Ethical Conduct confirming that they have complied with competition and antibribery rules and have not engaged in market manipulation. An auction monitor will be appointed to verify that the rules are followed.

22.3.4.2 Minimizing the scope for distortion DECC has included a number of measures in order to minimize the scope for distortion of the market. The government described them in its Consultation on Proposals for Implementation, as follows:67 (a) Pre-qualification is proposed to be mandatory so where plants opt out of the auction, demand can be lowered. (b) Participants’ capacity value is proposed to be administratively determined— preventing providers from understating the capacity value of their plants to withhold capacity from the auction. (c) There are proposed provisions to cancel or postpone the auction if it is undersubscribed. (d) Existing plants are proposed to be required to act as price takers in the auction unless they justify the need for a higher payment. (e) The level of supply in each auction and the identity of particular bidders are proposed to be concealed to mitigate risk of collusion. (f) The proposed demand curve will ensure the capacity price is less sensitive to the volume of capacity offered into the auction and so reduce incentives for participants to withhold capacity from the auction. (g) The proposed auction monitor will monitor the auction and provide validation that the auction followed the rules. (h) The proposed strong penalty regime for providers that fail to deliver energy when needed will help mitigate the risk of overpayment for unreliable plant. The government also separately mentions restrictions on the ability of capacity to declare itself non-operational and then re-enter the capacity market. With the Consultation on Proposals document,68 DECC published a Report on Capacity Market Gaming and Consistency Assessment prepared by Charles River Associates.69 It assessed the risks of gaming through strategies such as withholding capacity and collusion. It also considered the possibility of manipulation by DSR providers, in particular by the strategy of increasing load in the run-up to a call for DSR, and thereby appearing to reduce load but in reality only reducing it from an 66 Regulation (EU) 1227/2011 of the European Parliament and of the Council of 25 October 2011 on wholesale energy market integrity and transparency [2011] OJ L 326/1-16 (REMIT). 67 DECC, EMR: Consultation on Proposals (n 39) para 731. 68 DECC, EMR: Consultation on Proposals (n 39). 69 Charles River Associates, Capacity Market Gaming and Consistency Assessment—Final Report, CRA Project No D18985-00, September 2013, prepared for DECC (CRA Report).

380

United Kingdom

artificially high baseline. The Report pointed to the proceedings of the US Federal Energy Regulatory Commission (FERC) against Rumford Paper Company for manipulation of its DSR baseline as an illustration of the risks, albeit in that case not in the context of a capacity market. The Report concluded that the design of the GC capacity market provides substantial mitigation against the principal gaming risks. However, it also pointed out that no process is immune from risks. It highlighted the need for monitoring and review. Where insider trading or market manipulation in respect to the capacity market fall within the scope of REMIT, OFGEM has enforcement powers including the power to impose unlimited penalties, and can apply to the courts for injunctions and restitution orders.

22.4 European dimension 22.4.1 Acknowledging the EU context The government had claimed throughout the process of designing the capacity market that it had taken EU law obligations into account. For example, in the July 2011 White Paper,70 it expressed the view that the Electricity Market Reform project as a whole was consistent with EU energy policy. It also expressed the view that in principle the challenges faced by the GB market were best addressed through EU efforts. It also acknowledged the fact that EMR as a whole was subject to EU state aid approval, and sought clearance before the Regulations were adopted and the auctions started. The Commission gave state aid approval to the GB capacity market.71 The Commission commented that this was the ‘first time that it had assessed a capacity market under the new provisions on capacity markets in the new Environmental and Energy State Aid Guidelines.’ Generally, the GB capacity market seems to have been well received in Brussels: Commissioner Almunia said ‘the UK Capacity Market embraces the principles of technology neutrality and competitive bidding to ensure generation adequacy at the lowest possible cost for consumers, in line with EU state aid rules.’72

22.4.2 Exclusion of interconnected capacity A particularly interesting issue raised by the GB capacity market in the context of EU law is the treatment of non-GB generation. DECC initially envisaged that non-GB generation would be eligible to participate in the capacity market, although it recognized that there might be technical constraints. However, the government then identified a number of obstacles to the full participation of interconnected capacity. They included the difficulty of verifying the delivery of energy by overseas plants, of assessing the appropriate de-rating of interconnected plants and of ensuring that a plant 70

DECC, White Paper (n 15). Commission decision of 23 July 2014 in Case SA.35980 (2014/N-2) United Kingdom Electricity Market Reform—Capacity Market C(2014) 5083 final, [2014] OJ C/348 (UK capacity mechanism). 72 European Commission, Press release IP 14/865, 23 July 2014. 71

22.4 European dimension

381

participating in the GB capacity market does not also benefit from a support scheme in another Member State and thereby receive overcompensation. It therefore concluded that non-GB generation should not be able to participate in the 2014 auction. The government recognized that ‘excluding interconnection could also be troublesome for [its application to the Commission for State Aid clearance of the capacity market].’73 The Commission noted the exclusion in its clearance decision, but also noted the government’s commitment to enable participation of interconnected capacity in the 2015 auction.74 It also noted that as the capacity market is a remedy for ‘missing money’, addressing the missing money for GB capacity but not for interconnectors could lead to underinvestment in interconnection and overinvestment in potentially more expensive GB plants instead. The most significant obstacle seems to be the question of compatibility with the Target Model. As DECC explained:75 It is the objective of the Target Model to ensure that interconnector flows are determined implicitly by price differentials between the markets, rather than explicitly through procurement of physical transmission rights over the interconnector. This means that it is not possible to identify whether any particular plant has directly contributed to interconnector flows. Because of this it would be impossible to impose a CM penalty on a non-GB generator if it had delivered in its home market, but the interconnector from that market to GB was not delivering energy (and hence its capacity) to GB. Because of this it is difficult to see how the treatment of non-GB plant could be equitable with GB plant given that when a plant delivers in GB it clearly always primarily delivers its energy (and capacity) onto the GB system.

DECC therefore put forward a revised proposal that sought to address these concerns, allowing interconnectors to participate, although on slightly different terms to those on which domestic generation participates. The interconnector owner (rather than the interconnected generation) will be the capacity agreement holder. Essentially National Grid will apply a specific de-rating algorithm that will reflect the likelihood of the interconnector exporting to GB at times of system stress. The interconnector would be required to meet all other pre-qualification criteria, including the requirement for a UK corporate group member to enter into the capacity agreement. The interconnector would then participate in the auction as a price taker. It would then receive capacity payments but would be exposed to penalties if the load across the interconnector in a stress period was less than its load-following capacity obligation. It would be able to access the balancing services market in order to facilitate the delivery of energy during times of system stress.76 The government issued a consultation in September 2014.77 73 DECC (Capacity Market Expert Group), Interconnection and the Capacity Market, Meeting paper CMEG32.05 for 17 December 2013, available at https://www.gov.uk/government/policy-advisorygroups/capacity-market-emr-expert-group, accessed 1 February 2015. 74 UK capacity mechanism (n 71). See discussion in section 9.5. 75 DECC, October 2013 IA (n 19). 76 DECC, Interconnection and the Capacity Market (n 73). 77 DECC, Electricity Market Reform: Capacity Market Consultation—Consultation on Capacity Market supplementary design proposals and Transitional Arrangements, URN 14D/356, 25 September 2014.

382

United Kingdom

22.5 Conclusion The development of the capacity market has been a major policy exercise for the government and the GB energy sector, potentially affecting a very significant proportion of GB generation and DSR capacity. The first auction was expected to procure 48.6 GW of capacity. In comparison, total UK capacity at the end of 2013 was 85 GW. Although the capacity market will therefore take a large part of the GB energy market outside competitively determined prices, nevertheless the government expects the GB capacity market to result in a net increase in the average annual domestic electricity bill of only £2 over the period 2016 to 2030.78 If true, this will be a reflection of the lengthy and detailed analysis and consultation process conducted by the government. The Commission’s validation of the GB capacity market also means that it stands as a valuable precedent for other markets looking to implement similar mechanisms.

78

DECC, September 2014 IA (n 22).

Index NOTE: In order to avoid multiple repetitions and cross-referrals, many of the main entries are also to be found as sub-entries under the following countries:- Austria, Belgium, France, Germany, Greece, Italy, Netherlands, Norway, Poland, Spain, and United Kingdom. abuse of dominant position 189, 191–2, 194 Agency for Cooperation of Energy Regulators (ACER) 10, 83–4, 160 Opinion on Capacity Markets 33 Report on Capacity Mechanisms and the EU Internal Electricity Market 32–9, 48, 202, 204, 223 cross-border participation 37–8 energy-only markets 33–4 impact of capacity mechanisms: design and distortions 35–7 recommendations 38–9 Albania 289 allocation step 103 Almunia, J. 161, 380–1 ancillary services 173, 188 Annualized Forward Capacity Payment 135 anticompetitive practices 189–200 antitrust law 182–200 abuse of dominant position 189, 191–2, 194 anticompetitive practices 189–200 auction design 196–8 bid rigging 185 capacity auctions 185–6 capacity obligations (with markets for certificates) 185–6, 188 capacity payments 185–6 capacity withholding 185, 191–2 cartels 187, 195 collusion 185, 189, 192–3, 194 enforcement in energy sector 183–5 enforcement limitations 193–6 financial market manipulation 186 gaming 189–93 insider trading 186 joint bidding strategies 193 market monitoring 196–8 price-fixing 187 relevant market 187–9 reliability options 185–6, 188 REMIT 198–9 strategic bidding 189 strategic reserve 185–6, 188 transmission system operator (TSO) 185, 187, 199–200 typology of issues 185–7 Argentina 16, 123 Arriaga, P. 53 Arrow, K. J. 145n adequacy assessment 48–51 auctions 3, 12, 13, 185–6 antitrust issues 189–99 buying side 121–5, 138 capacity payments 16

centralized 123–5, 129, 183 coordination rationale and potential 81, 84 cross-border participation 100, 101, 107, 109–10 decentralized 123–5 descending-clock auction 191–2 design elements 121, 196–198 explicit cross-border participation with congestion on interconnections 103, 105n long-term 119–139 multi-round 193 pay-as-clear 190–2 periodic capacity 193 selling side of auction 125, 126 social efficiency of cross-border participation from national perspective 107, 109–10 state aid rules and funding 165, 177, 178, 179 system adequacy problems 121 Australia 11 Austria 85, 227–40, 242–3, 250, 255 additional support scheme 236–7 assessment of proposed capacity mechanism under EU law 238–40 auctions 229 Austrian Power Grid (APG) 229–33 base-load 234 and Bosnia 229 Branchenorganisation 236, 239 and Bulgaria 229 Chamber of Commerce (WKO) 236 Clearing and Settlement Agency 232 coal-based power 227 cogeneration model (Branchenregeln) 236–40 cogeneration points (KWK-Points) 236–40 combined cycle gas turbine (CCGT) power plants 235 Combined Heat and Power (CHP) certificate system 238 congestion 229 Constitution 239–40 coverage of power consumption by domestic generation 234 and Croatia 229 and Czech Republic 229, 231 devaluation of assets 233 Distribution System Operators (DSOs) 232 E-Control 232–3, 235–6, 240 eco-electricity subsidy 228 EEX 235 Energie AG 228 Energie Steiermark AG 228 Energy Efficiency Package (draft) 236 Energy Exchange Austria (EXAA) 228, 235 Energy Intervention Powers Act (2012) 232 energy-only market 235–7

384 Austria (cont.) European dimension 237–40 EVN AG 228 Federal Ministry of Economy, Family and Youth 232 Federal Ministry of Science, Research and Economy 236 feed-in tariffs 228, 231–2 and France 229 free movement of goods 212 fundamental right to property 239–40 gas-fired plants 227 generation adequacy 233–5 geothermal power 228 and Germany 229, 231, 233, 235, 237, 286 green energy 228 Green Energy Act (2012) 231–2 hydropower 227–8, 229 Illwerke AG 228 interconnection/interconnectors 229 and Italy 229 KELAG 227, 228 KPG 236–7, 239–40 Legal and Constitutional Service of the Federal Chancellery 239–40 Linz AG 228 Main Committee of the National Council 232 market characteristics 227–31 market price 239 mothballing 233–5 and Netherlands 238 network development plan (2013) 237 nitrogen oxides (NOx) trading scheme 238 peak-load 234 photovoltaic/solar power 227–8, 235, 237 and Poland 231 proposed support scheme 236–7 public ownership 227 regulatory framework 231–3 renewable generation sources (RES) 228, 231, 235, 237–40 and Romania 229 Salzburg AG 228 and Serbia 229 and Slovenia 229 social efficiency of cross-border participation from EU perspective 113n and Spain 364 state aid 170, 238–9 state ownership 227 subsidization 240 and Switzerland 229 thermal generation 227–8, 233–5, 237, 240 Third Party Access 233 Tiroler Wasserkraft AG 227 TIWAG-Tiroler Wasserkraft AG 228 transparency unit 236, 239–40 Verbund AG 227, 228, 233 wholesale market/price 228, 231, 235 Wien Energie 228, 233 wind generation 227–8, 231, 235, 237 Worarlberger Illwerke 227 Working Program (2018) 237–8

Index Baker, P. 104 balancing market 116, 173, 186, 188 critical period 132 cross-border participation 100 system adequacy problems 139 balancing test 174, 179 Barnard, C. 213 Barroso, L. A. 137n base-load capacity 140, 153, 154 capacity payments and market efficiency 149–50 optimal generation mix 142, 143–7 price caps and investment level distortion 148 renewable energy and demand participation 151–2 Bavaria 5–6 Belgium 85, 241–55, 318 Allegro project 247 auction 250 and Austria 255 capacity tendering 255 Central and Western European (CWE) region 242–3, 250 coal-fired plants 242 combined cycle gas turbine (CCGT) power plants 247–9, 253 Commission Communication (November 2013) 26 Commission Generation Adequacy Staff Working Document (SWD) 255 day-ahead market 250 demand response 246 Elia (TSO) 241–3, 245–6, 249–52 Energy and Environmental Guidelines (EEAG) 2014–2020 253 Energy Policy Plan 246–7, 249–50 European dimension 253–5 existing capacity mechanism 245–6 explicit cross-border participation with congestion on interconnections 102n Federaal Planbureau/Bureau féderal du Plan 242 Federal Energy Administration 242, 248–53 and Finland 254 Flanders 207–8, 218 flexibility of generation 244–6 Forecast Study 242–3 and France 242n, 244, 246n, 247, 254–5 free movement rules 255 gas-fired plants 242 generation adequacy 243–4, 245, 255 and Germany 246n, 254–5 grid stability 243, 245 harmonization 250 hydropower 242 interconnection/interconnectors 246, 247 intraday market 250 and Italy 254 Loss-of-Load Expectation (LOLE) 251 market characteristics 241–2 market coupling 242 Ministerial Decision (2013) 247–8 ‘missing money’ problem 244 mothballing and/or closures 243–4, 249–54 NEMO project 247 and Netherlands 242n, 244, 246n, 254–5

Index nuclear phase-out 242, 243–4, 246, 255 open cycle gas turbines (OCGTs) 247–9, 253 overcompensation 254 peak-load 246 photovoltaic/solar power 244 and Portugal 254 primary reserve 245 R4 reserve 245–6 Regulatory Commission for Electricity and Gas (CREG, energy regulator) 241–7, 250–2, 254 regulatory framework 242–3 reserve capacity 245 Royal Decree (2013) 247–9 secondary reserve 245 Secretary of State for energy (energy minister) 243 security of supply 243–6, 252 self-sufficient capacity mechanism, cost of 44 semi-base-load 246 and Spain 254 state tendering for new capacities 246–9, 253–4 strategic reserve 11, 246, 249–55 and Sweden 254 and Switzerland 255 tertiary reserve 245 thermal generation 244 Tihange 1 power plant 241, 243–4 transmission system operator (TSO) 242, 245, 249–50, 251–2 see also Elia Trilateral Market Coupling (Belgium, France and Netherlands) 242n and United Kingdom 246n, 254–5 wind generation 244 Bertsch, J. 41–3 bid prices 105, 108 bid rigging 185 bid/ask spread 68n bidding: competitive 174 joint 193 bilateral agreements (or contracts) 13, 54, 125, 185, 188 bilateral cross-border trading 98 bilateral obligations 100 billing systems 68 biomass 158 blackouts 6, 32, 65, 106, 115, 130 Boiteux, M. 142n Bortoni, G. 310–11 Bosnia 229 Brazil: A1 auction (1 year lag for existing generating units) 128, 129 A3 auction (3-year lag for thermal generation) 128 A5 auction (hydro projects) 127n, 128, 129 biomass 127n buying side of auction 122, 123–4 capacity auctions 13 capacity payments 16 critical period 131 El Niño Southern Oscillation (ENSO) 129

385

reliability in capacity- and energy-constrained systems 131 selling side of auction 126 wind generation 127n brownouts 65, 68 Buchan, D. 5n Buehler, S. 147n Bulgaria 229, 289 buying side of auction 121–5, 138 call options 101n capacity adequacy 103, 110–11, 115, 117, 967 Capacity Allocation and Congestion Management (CACM) Guideline (draft) 83 capacity costs 62 capacity expansion level 119 capacity leakage 86, 89 capacity margin criterion 89 capacity market warning 133 capacity obligations 12, 13–14, 37, 69, 81, 84, 100–1, 185–6, 188 capacity payments 12, 81, 87, 126, 151, 185–6 capacity premium 15 capacity rights 99–100, 103 capacity withholding 185, 191–2 capital costs 143, 151, 153 cartels 187, 195 central agents 69 central procurement 72–3 Central and Western Europe (CWE) region (Austria, Belgium, France, Germany, Luxembourg, Netherlands) 85, 242–3, 250, 285, 317 centralization 10, 69, 84, 100, 123–5, 138 Certificate of Ethical Conduct 198 certificates 13, 125 green 208, 222 market 13, 185 certification 89, 93, 199 Chile: buying side of auction 122, 123 capacity payments 16 critical period 131 indexation formulas 137 selling side of auction 126 system adequacy problems 119 tradable quantities constraints 135 classification of capacity mechanisms 70 clearing price 38n, 108, 117, 122, 191 coal-fired power 6, 158 collective goods 96, 102, 103 see also adequacy; reliability collusion 185, 189, 192–3, 194 Colombia: buying side of auction 123 capacity auctions 13, 136 critical period 131–2 firm energy certificates (ENFICC) 136 Firm Energy Obligation (OEF) 128, 131, 136 GPPS 128n hydro generation 128 lag period (lead time) 127 Platts US Gulf Coast Residual Fuel No 6 1.0 per cent sulfur fuel oil 132

386

Index

Colombia: (cont.) Regulatory Commission for Electricity and Gas (CREG) 136 reliability charge mechanism 128 reliability options 15 thermal generation 128 Wholesale Electricity Market Monitoring Committee 1334 combined cycle gas turbine (CCGT) power plants 5–6, 8, 14, 44, 178, 190n, 194, 197 Commission Communication on A policy framework for climate and energy in the period from 2020-2030 (2014) 22–30, 31, 161, 286 Commission Communication on Delivering the internal electricity market and making the most of public intervention (November 2013) 19, 22–30, 92–3, 311, 318 cross-border participation 29–30 demand side response 27–8 environmental protection 30 recommended steps, questionability of 24–6 regulatory failures, removal of 28–9 state aid control 157, 160 technological neutrality 30 Commission Communication on EU Guidelines for the application of State aid rules in relation to the rapid deployment of broadband networks (December 2013) 162 Commission Communication on Generation adequacy, capacity mechanisms and the internal market in electricity (2012) 21–2, 332–3 Commission Communication on Guidelines on State aid to airports and airlines (2014) 162 Commission Communication on Progress towards completing the Internal Energy Market (2014) 181, 223 Commission Communication on Smart Grids: from innovation to deployment (2011) 9 Commission Communication on State aid to promote risk finance investments (2014) 162 Commission Communication on Stranded costs (2001) 159–60, 171 Commission draft Notice on the notion of state aid (2014) 167, 169 Commission Service of General Economic Interest (SGEI) Package 163 Commission Staff Working Document on Generation Adequacy in the internal electricity market - guidance on public interventions (Generation Adequacy SWD) 19, 24–5, 28–30, 31, 76, 96, 110–11 Belgium 255 Germany 275 Greece 296 Italy 312 common interest objective 177–8 compatibility criteria 175–6 compensation 179 competitive market and generation mix, optimal 145–7

congestion 84, 187 cross-border participation 98, 105–6, 117–18 social efficiency of cross-border participation from EU perspective 110–11, 116 social efficiency of cross-border participation from national perspective 107 contract duration 128–30, 138, 186 Contracting Parties of the Energy Community 312 contracts for difference (CfDs) 14, 178, 180 coordination: cross-border capacity 90–3 day-ahead market 85 different designs 80–1 different drivers 81 distortions associated with uncoordinated capacity mechanisms 85–6 Energy and Environmental Guidelines (EEAG) 2014–2020 86–8 forward market 85 intraday market 85 key issues 82 local specificities 82 main objectives 82 power markets integration 83–5 rationale and potential 79–94 see also coordination framework of national capacity mechanisms coordination framework of national capacity mechanisms 48–54 assessment of adequacy needs and cross-border resources 48–51 rights over system resources during times of extreme scarcity 52–4 risk allocation and remuneration of cross-border resources 51–2 cost of new entry (CONE) 374 net 190–1, 194, 196, 198, 375, 377 cost-benefit analysis 9 cost-of-service scheme 126 costs: capacity 62 capital 143, 153 fixed 62, 80, 86 fuel 143 going-forward 194n, 198 investment 62, 159 marginal 5, 70, 145–8, 150, 153 operating 143 opportunity 43 peaking unit construction 113 production 143 short run marginal (SRMC) 86, 133 stranded 74, 159–60, 171 sunk capital 151 sunk fixed 126 total 149–50 total production 146–8 transaction 71 variable 61–2, 143, 146 Council of European Energy Regulators (CEER) 5n, 21, 48–9 Cramton, P. 9n Creti, A. 86 critical period 131–3, 139

Index Croatia 229 cross-border capacity 90–2 cross-border contribution 90 cross-border distortions 35 cross-border exchange prerequisites 92–3 cross-border forward markets 83n cross-border participation 95–118 additional capacity contributing to domestic capacity 107–8 adequacy policy, lowering cost of 108–10 Agency for Cooperation of Energy Regulators (ACER) Report on Capacity Mechanisms and the EU Internal Electricity Market 37–8 capacity rights and reliability rights 99–100 Commission Communication (November 2013) 29–30 distortive effects 114–16 explicit cross-border participation 49–50, 91, 93, 98–9, 106, 108–14, 117 with congestion 102–6, 115–16 without congestion 115 implicit cross-border participation 49–50, 91, 93, 98, 106, 108–11, 112–14, 116–18 with congestion 115–16 without congestion 115 market coupling 104–6, 116–17 physical and economic fundamentals of capacity rights trade 102–4 simplifying hypotheses 101–2 social efficiency 106–10 from EU perspective 110–17 from national perspective 106–10 solidarity principle 105–6 cross-border resources, measurement of 48–51 cross-border support 75–6 curtailment levels/ratios 52–4 customs duties, charges and taxes 203 Czech Republic 45, 229, 231, 286, 336 day-ahead auction/market 84–5, 100, 104–5, 116, 132, 139 de minimis rule 187 De Nooij, M. 144n deadweight loss 147–9, 153 decentralized agents 69 decentralized auction 124–5 decentralized capacity mechanisms 10, 123 demand 7–8, 81 captive 122 free 122–3 maximal 145 participation and renewable energy 150–3 rationing 148, 151 response bids 138 demand side response (DSR) 7, 8–10, 27–8, 133, 138 Denmark 364 Energy Agency 332 and Norway 323, 326, 327, 331–2 descending-clock auction 191–2 diesel plants 6 disconnection 27 discount rates 129 discrimination, arbitrary 213–14

387

dispatch efficiency 84, 86 dispatchable plants 60 Distribution System Operators (DSOs) 85, 163n, 173–4, 178, 181, 188, 197 dynamic efficiency of power markets 85 EAG 2008 171 economic advantage 164 economies of scale 124, 138 EDF 27 efficiency 74, 83, 153 see also social efficiency Electricity Coordination Group 31, 161 Electricity Directive (2003) 16–17 Electricity Directive (2009) 19 electricity margins, low, as justification for capacity mechanism 62–4 energy efficiency bids 138 Energy and Environmental Guidelines (EEAG) 2014–2020 86–8, 171–9, 181 Austria 239–40 Belgium 253 common assessment principles 172–5 coordination rationale and potential 79, 91, 94 cross-border participation 90, 95 distortions of competition and effect on trade and positive balance 88 draft 22 Greece 297 Italy 311, 312, 313 objective of common interest 87 Poland 347–9 State Aid Modernization (SAM) 161–2 state aid rules and funding 86–8, 157–8, 162–6 United Kingdom 175 energy market design 36, 59–78, 121 challenges 75–6 Commission guidance 76 European dimension 78 existing market design: reform 77 flaws 63–4, 71 issues 75–6 options and market failures 69–72 reform with capacity mechanism 77–8 reforms 77–8 see also energy-only market: sustainability; policy options energy payments 60 energy prices 43, 70 energy sector inquiry 184 energy-only market: design flaws 63–4 functioning 61–2 low electricity margins as justification for capacity mechanism 62–4 market environment, evolving 63 market failures 64–7 market power, threat of in peaking periods 67 ‘missing money’ problem 67 political intervention 63 security of supply as ‘public good’ 65–6 sustainability 60–7 underinvestment due to prohibitive price risk 66–7

388

Index

energy-only market 15, 33–4 Austria 235–7 Germany 276–85 Netherlands 318–20 Norway 325–31 Poland 342–7 entry barriers 124–5 environmental protection 30, 214, 218–23, 307, 338 equity principle of tariff design 124 Estonia 326, 327 Estonian case (Aid for Capacity Payments for Oil-Shale Fuelled Electricity Production (2011)) 20 EU Emission Trading Scheme (ETS) 63, 265, 269 EU Target Model 83–4, 104, 291, 299 Euphemia algorithm 53 European Competition Network 185 European dimension: Austria 237–40 Belgium 253–5 France 264–70 Germany 285–6 Greece 297–301 Italy 310–13 Netherlands 318–20 Norway 331–4 Poland 347–9 Spain 363–4 United Kingdom 380–1 European Economic Area (EEA) 207–8, 312, 331 European Energy Exchange (EEX) 285 European Network of Transmission System Operators for Electricity (ENTSO-E) 49, 106, 118, 160, 172, 178 coordination rationale and potential 83, 90 generation adequacy assessment 45, 87, 89 Greece 292, 296 Netherlands 317 Scenario Outlook and Adequacy Forecast 20142030 25, 234 Ten-Year Network Development Plan (TYNDP) 89 European Parliament’s Industry, Research and Energy (ITRE) Committee 33 excess capacity and self-sufficiency 45–7 existing plants, role of 125–6 export restrictions 88 Fabra, N. 86 financial market manipulation 186 financial settlements 199 financial warranties 136, 137–8 Finland 11, 21n, 88n, 204–6, 220, 254 and Denmark 323, 326, 327, 331, 332 Finon, D. 54, 86 First Welfare Theorem 145n fixed costs 62, 80, 86, 126, 142 flexibility of capacity 3–4, 6, 8, 43–4, 47, 55, 68, 81, 132 flow-based cross-border capacity calculation 84 foreign participation 39 forward capacity markets 83n, 85, 100, 104, 111–13 forward commitments 100

Framework Guidelines 83–5 France 6, 229, 256–70, 303, 335 antitrust issues 267–8 ARENH 259, 261 assessment of adequacy needs and cross-border resources 50 and Belgium 242n, 244, 246n, 247, 254–5 bonus-malus system 262–3, 269–70 Capacity Decree 261–3, 268 capacity obligations 14, 258, 262 CDC 170 certificate market 258, 260, 262–3, 267–70 certification and control of generation or demand response capacities 263 competition authority 27, 265–8 coordination rationale and potential 81–2, 84, 89, 90 CRE (energy regulator) 170, 258 critical period 133 cross-border participation 97 demand-side response 267 deposit and consignment office 27 Distribution System Operators (DSOs) 261, 263 EDF 97n, 256, 259, 261, 267 electro-intensive consortium (Exeltium) 261 energy market design 59 EU emission trading scheme (ETS) 265, 269 European dimension 264–70 free movement rules 264–7 generation adequacy 22, 257–8, 261, 269–70 interconnection/interconnectors 266 internal energy market directives 264–7 Minister of Ecology, Sustainable Development and Energy (energy minister) 97n, 258 Ministerial Decrees (2009) 257 multi-year plan of investment (MPI) 257 multiplying coefficient 50 NEMO Law 259–61, 266, 268 Nome Law 14n non-discrimination 265–6 nuclear energy 259 over-the-counter (OTC) transactions 268 as part of Central and Western Europe (CWE) 85, 242–3, 250 price spikes 258 proportionality test 270 public service obligation (PSO) 265 reciprocity clauses 267 RTE (TSO) 50, 97n, 125, 245, 257 security of supply 259, 260–1, 264–7 self-sufficiency and excess capacity 45 self-sufficient capacity mechanism, cost of 44 Sido-Poignant Report (2010) 256–8 and Spain 352 state aid rules 268–70 strategic reserve 267 system adequacy problems 120 technical rules 264 transmission system operator (TSO) 258, 260–4, 266, 268, 270 see also RTE transparency 265 and United States 259 wholesale access 259 working group 257–8

Index fraud 222 free movement of goods 201–23 discrimination, arbitrary 213–14 energy as goods 202–3 environmental protection 218–23 justification 210–23 prohibition of customs duties and charges of equivalent effect on imports and exports 203–4 prohibition of discriminatory internal taxation 204–6 prohibition of quantitative restrictions and measures having equivalent effect on exports 209–10 prohibition of quantitative restrictions and measures having equivalent effect on imports 206–9 proportionality 211–13 public policy and public security 214–18 restrictions 203–10 and state aid 203 free-riding 122–3 freezing during scarcity 54 fuel costs 143 funding capacity mechanisms and state aid rules 162–71 cumulative conditions and effects-based test 163–4 economic advantage 164 public service obligations (PSOs), compensation for 164–6 state resources, financing from 166–70 gaming 189–93 gas prices 63 gas-fired plants 158 generation adequacy 40, 55 Agency for Cooperation of Energy Regulators (ACER) Report 36, 38, 39 assessment of adequacy needs 48–51 coordination rationale and potential 80 critical period 132 cross-border participation 99, 102–3 cross-border resources, measurement of 49 Energy and Environmental Guidelines (EEAG) 2014–2020 172, 173 -oriented payment 126 policies and impact on energy markets 47 policies, lowering cost of 108–10 problems 121 remuneration of flexibility 42–3 rights over system resources during extreme scarcity 53–4 risk allocation and remuneration of cross-border resources 52 self-sufficiency and excess capacity 47 target 111 generation mix 41 optimal 141–7 geographic market 187–8 Germany 228, 246n, 254–5, 271–87 alternative measures 281–5 antitrust law 189n

389 Association of Energy and Water Industries (BDEW) 283–5 Association of Municipal Enterprises (VKU) 283 Association for Renewable Energies (BEE) 284 and Austria 229, 231, 233, 235, 237, 286 balancing group managers 283–4 base-load 272 biomass 274 capacity auctions 13 carbon dioxide certificates 272 carbon prices 282 Central and Western Europe (CWE) region 85, 242–3, 250, 285 Commission Communication (November 2013) 26 coordination rationale and potential 81–2, 89 and Czech Republic 286 day-ahead market 284 decentralized capacity mechanism 283–4, 285 electricity market 2.0 287 Electricity Market for the Energy Transition (green paper) 271, 282, 287 EnBW 195 Energy Concept (2010) (Federal Government) 271–2 Energy Industry Act (EnWG) 280–1 energy market design 60 Energy Package 272 Energy Transition (Energiewende) 271, 281, 283, 287 energy-only market 63, 276–85 enforcement 281 EPEX Spot 284 European dimension 285–6 European Energy Exchange (EEX) 285 Federal Cartel Office (FCO) 195–6 Federal Council 281 Federal Government 271–2, 275, 280–1, 286 Federal Ministry of Economic Affairs and Energy (BMWi)(Energy Ministry) 15, 271, 272, 275, 282, 284, 287 Federal Network Agency for Electricity, Gas, Telecommunications, and Railway (BNetzA) (energy regulator) 196n, 274–5, 276, 278–81, 284–5 and France 276n, 286 free movement of goods 207, 209, 219–20 fuel prerogatives of large gas-fired power plants 280 gas-based power plants 272, 280 generation adequacy 275–6 ‘grand coalition’ 273, 286 greenhouse gas emissions, reduction of 272 inflexible system 275 Internal Energy Market 286 and Italy 286 maintenance expenses 280 marginal costs 272 market characteristics 273–4 market decoupling 285 market prices 273 market-related measures 277

390

Index

Germany (cont.) Ministry for the Environment, Nature Conservation and Nuclear Safety 283 mothballing and closures 279, 280 and Netherlands 276n, 318 network reserve 271, 277–9, 282, 287 no final shut down of system relevant generation and storage units 279–80 north-south imbalance 272–3 and Norway 326, 333–4 notification requirement and twelve-month moratorium on plant closures 279 nuclear phase-out 271–2, 273, 275, 281 one-off tendering procedure 286 Ordinance on Reserve Power Plants (ResKV) 277–8, 285–6 oversupply 271 peak-load demand 272, 284 Pentalateral Forum 286 photovoltaic/solar power 272, 274 and Poland 335, 348 policy options 69n power exchange (EEX) 195 redispatch 277, 284–5 regulatory framework 274–5 remuneration of flexibility 41 renewable generation sources (RES) 271–4, 276, 287 RWE 195 security of supply 275–6, 284–6 self-sufficiency and excess capacity 45 self-sufficient capacity mechanism, cost of 44 shortage 271 shortage price sellers 284 social efficiency of cross-border participation from EU perspective 113n special certificates (VSN) 283–4 state aid 169–70, 176, 180, 272, 287 strategic reserve 11, 271, 280, 281, 283, 284–5, 286 and Switzerland 276n, 278, 286 system adequacy problems 120 system relevance 279 transmission grids 274 transmission system operator (TSO) 276–7, 278–81, 284, 285, 286 Vattenfall 195 White Paper (proposed for 2015) 282, 287 wind generation 274 Germany as part of Central and Western Europe (CWE) region 85, 242–3, 250 going-forward costs 194n, 198 Gottstein, M. 104 Gräper, F. 254n Greece 288–301, 356 ADMIE (Hellenic transmission system operator (TSO)) 288–9, 296 and Albania 289 and Bulgaria 289 Capacity Adequacy Study 290 Capacity Assurance Mechanism (CAM) 289–90, 293–4, 296, 298, 299–301 Transitional 289–90, 294–5, 297–9 Capacity Availability Contracts (CACs) 294

Capacity Availability Tickets (CATs) 293–4, 299 Register 294–5, 299, 301 capacity obligations 14 capacity payments 15–16 certificates 293 cogeneration 292 Commission’s Generation Adequacy SWD 296 contracts for difference (CfDs) 293 coordination rationale and potential 84 Council of State 297 day-ahead pool model 290 DEDDIE (Hellenic Distribution Network Operator (DSO)) 288–9 Development Plan (10 year) 291 Distribution System Operators (DSOs) 288–9 Energy and Environmental Guidelines (EEAG) 2014–2020 297 EU Target Model 291, 299 European Courts 298 European dimension 297–301 European Network of Transmission System Operators for Electricity (ENTSO-E) 292, 296 and Former Yugoslavian Republic of Macedonia (FYROM) 289 free movement of goods 216–18, 296, 299–300 generation adequacy 291–3, 296, 300 Generation Adequacy Study 296 generation capacity 297 Generation Unit Register 299–300 Grid and Market Operation Code (Grid Code) 290, 293–4, 299 hydropower 292, 293, 295 interconnected system (mainland) 289, 291–2 and Italy 289 lignite 292, 295 load representatives 294, 298 market characteristics 288–90 market distortions 294–5 Memorandum of Understanding 301 non-discrimination 300 non-interconnected network 289 peak demand 292 proportionality principle 296 Public Power Corporation (PPC) 288, 291, 295, 297 public service obligation (PSO) 298 Regulatory Authority of Energy (RAE) (energy regulator) 290, 294–7 Annual Report (2003) 293 regulatory framework 290–1 renewable generation sources (RES) 289, 291–2, 296, 297 intermittent 292 sector-specific provisions 300–1 security of supply 293, 300 selectivity 299 self-sufficiency and excess capacity 45 state aid rules 297–9 strategic reserve 297 Supreme Administrative Court 297 thermal generation 292 Third Energy Package 290

Index transmission system operator (TSO) 288–90, 294, 296, 298 transparency 300 and Turkey 289 wholesale prices 291–2 green certificates 208, 222 grid code 43 group exemptions 187 harmonization 48–9, 83, 85, 89–90, 91, 93–4, 96 hedging the end-user default tariff price 122 Herfindahl-Hirschman Index (HHI) 193n HIDENFICC computational model 136 Hogan, W.W. 141n hourly capacity margins 45 Hungary 336 hybrid designs 185, 188 hydropower 113, 117, 129 reliability in capacity- and energy-constrained systems 131 state aid control: policy evolution 158 strategic reserve 11 tradable quantities constraints 136 impact assessment 35–7, 39, 71 import quotas 206 incentive effect 179 indexation formulas 137 indirect control test 167 insider trading 186 interconnection/interconnectors 25–6, 29–30, 41, 50–1, 81, 86, 174, 178 International Energy Agency (IEA) 22n international price of fuels 137 intraday market 83n, 84–5, 100, 104–5, 116, 132, 173 investment: costs 62 costs, non-returned 159 decisions 35–6 levels 146–9, 151–2 long run 147–9, 151 lumpy 65 short run 147–8, 151, 153 Ireland 356 capacity payments 15–16 coordination rationale and potential 84–5 free movement of goods 207, 214–17 grid code 43 optimal generation mix 144n self-sufficiency and excess capacity 45 Single Electricity Market (SEM) 50 system adequacy problems 119 Irish CADA case (2003) 17–18, 20 Italy 229, 254, 289, 302–13, 356 Acquirente Unico 28–9 AEEG (energy regulator) 129, 304–8, 310 Agency for Cooperation of Energy Regulators (ACER) 311 assessment of adequacy needs and cross-border resources 50 assessment of capacity mechanisms under EU law 311–13 auctions 308

391 ‘Barcelona target’ 311 base load 303 ‘Bersani Decree’ 306 blackouts 311 capacity payments 15–16, 306–7, 311 central auctioning 15 combined cycle gas turbine (CCGT) power plants 303 Commission Generation Adequacy Staff Working Document (SWD) 312 contract duration 129–30 contribution of major groups to domestic gross production 304 contribution of major groups to production of electricity for consumption 304 coordination rationale and potential 81–2, 84–5 day-ahead market 303, 307, 309 eligibility criteria 309 Enel 307, 352 Energia Concorrente 308 Energy and Environmental Guidelines (EEAG) 2014-2020 311, 312, 313 energy market design 59 enforcement powers 309–10 environmental legislation 307 European dimension 310–13 exercise price 308–9 existing capacity mechanism 306–7 final capacity mechanism 307–10 and France 303 free movement of goods 202, 313 geographical zones 302 GME (market operator) 302 Government 310 GSE (previously GRTN) 306 imports 303, 311 information asymmetry 307 interconnection/interconnectors 311 internal market dimension 311–12 macro-zone 302n marginal pricing 302 market characteristics 302–4 market failure 307 market pool 302 merit order 302 Ministry of Economic Development 304–5 National Strategy for Energy 310 net production by source (2012) 305 price zones 302–3 reference price 309 regulatory failures 307 regulatory framework 304–6 reliability options 309 renewable generation sources (RES) 303, 307–8, 312–13 sanctions 310 and Switzerland 303, 312–13 system adequacy problems 120 Terna 306–10 Framework Discipline 309 thermal generation 303 Third Energy Package 306 transmission system operator (TSO) 309–10 virtual zones 302

392

Index

Italy (cont.) weighted average of zonal prices (PUN) 302–3 wholesale prices 302–3 Japan 19n joint bidding strategies 193 Jones, C. 254n Joskow, P.L. 4n, 141n, 145n Kennedy, S. 151n Kilian, K. 340n lag period (lead time) 127–8, 138, 197 Lange, O. 145n Latin America 13n, 60, 127, 131, 189 see also Argentina; Brazil; Chile; Colombia; Panama; Peru Latvia 26 Latvian Case (Tender for Aid for New Electricity Generation Capacity (2010)) 19–20 Leahy, E. 144n Lerner, A.P. 145n lignite 292, 295, 335, 338–40 Lithuania 326, 327, 348 LNG terminal 166, 168 Loss-of-Load Expectation (LOLE) 36n, 89, 101 low-carbon 159 Luxembourg 85, 242–3, 250 Macedonia 289 mandatory requirements 214, 218 marginal costs 5, 70, 145–8, 150, 153 short-run 86, 133 marginal pricing 68, 80 market coupling: algorithm 55 assessment of adequacy needs and cross-border resources 50 coordination rationale and potential 83–4 cross-border participation 104–5 day-ahead 83n Energy and Environmental Guidelines (EEAG) 2014–2020 173 explicit cross-border participation with congestion on interconnections 106 implicit cross-border participation 116–17 rights over system resources during extreme scarcity 52–3, 54 social efficiency of cross-border participation from national perspective 107–8 market distortions 35, 52, 70, 85–6, 88, 114–16 market efficiency and capacity payments 149–50 market equilibrium 145–6 market failures 61, 64–7, 69–72, 80, 87, 140n–1n, 154 market liquidity 68 market makers 68 market power 67, 140n–1n market prices 61, 132–3, 145 market risks 63 market types see balancing market; day-ahead auction/market; intraday market; spot market; wholesale market market-based designs 84 market-clearing price 38n, 108, 117, 122, 191

market-sharing agreements 184 market-wide capacity mechanisms 10 Maurer, L. T. A. 137n merit order 3, 5, 34, 108, 153 metering systems 68 see also smart meters ‘missing money’ problem 4–5, 6, 8, 26, 153 coordination rationale and potential 80 Electricity Directive (2009) 19 energy-only market sustainability 65 price caps and investment level distortion 147, 148 regulatory intervention, threat of 67 renewable energy and demand participation 151 RES Directive (2009) 21 see also state aid control and ‘missing money’ problem Morocco 352 mothballing or closure of plants 6, 151 multi-round capacity auctions 193 multilateral agreements 54 multilateral regulatory framework 52 mutual insurance principle 55 mutual recognition principle 222 national competition authorities (NCAs) 185 National Energy Plans (NEPs) 31 national regulatory authorities (NRAs) 84 necessity test 184 Netherlands 238, 276n, 314–20 ACM (competition authority) 314, 317, 319–20 and Belgium 242n, 244, 246n, 254–5, 318 Central and Western Europe (CWE) region 85, 242–3, 250, 317 coal-fired plants (decommissioned) 316, 317 combined heat and power (CHP) plants 316 Commission Communication November 2013 318 coordination rationale and potential 89 demand-side response (DSR) 318 energy-only market 318–20 E.ON 316 European dimension 318–20 European Network of Transmission System Operators for Electricity (ENTSO-E) Report 317 gas-fired power plants 314, 316, 318 GDF/Suez 316 generation adequacy 22, 316–18 and Germany 318 HVC 316 installed capacity per generation source 316 interconnection/interconnectors 318 market characteristics 314 Minister of Economic Affairs 315, 317, 319 Monitoring Report on the Security of Electricity and Gas Supply 315–17 mothballing or decommissioning 318 National Energy Agreement 317 Northland Power 316 and Norway 317, 323, 326 nuclear power 316 optimal generation mix 144n overcapacity 314 peak demand 314, 317

Index photovoltaic/solar power 319 regulatory framework 314–15 renewable generation sources (RES) 314, 316, 319 reserve capacity auctions 315 RWE/Essent 316, 318 ‘second-best’ option 318, 320 security of supply 315, 318–20 self-sufficiency and excess capacity 45 SEP (transmission system operator (TSO)) 30 Siemens 316 social efficiency of cross-border participation from EU perspective 113n TenneT (transmission system operator (TSO)) 245, 315 thermal generation 316 transmission system operator (TSO) 30, 245, 315 and UK 317 Van Oord 316 Vattenfall/Nuon 316 wholesale market 315, 316, 318, 319 wind generation 316–17, 319 Network Codes 83–5 network modelling 49 New Zealand 11 Newell, S. A. 4n no double-counting 51 nomination step 103n non-discrimination 47, 88 non-dispatchable plants 60n non-tariff barriers 203, 206 Nordic countries 85, 89, 91 see also Denmark; Finland; Norway; Sweden North America 123, 124, 130, 138 see also United States Norway 317, 321–34 acknowledgement of Nordic context 331–2 area licence 324 assessment of capacity mechanisms under EU law 332–4 bidding areas 326–7 bilateral contracts 327 central transmission grid 330 certificates market 328–31 coal power 322 common Nordic electricity market 326–8 Common Plan for Water Systems 324 Competition Authority 323, 330, 331 Conservation Plans 324 contracts for difference (CfDs) 327 cross-border trade 326 daily reference price 327 day-ahead market 327 de-centralization 326 and Denmark 323, 326, 327, 331–2 deregulation 325 Distribution System Operators (DSOs) 324 distribution/electricity grid 330 Elbas 327 Elspot 327 energy-only market 325–31 enforcement of energy market regulation 330–1 and Estonia 326, 327

393 European dimension 331–4 facility licence 324 Financial Supervisory Authority 323, 330, 331 and Finland 323, 326, 327, 331, 332 forward contracts 327 fossil fuels 332 free movement of goods 207–8 future contracts 327 gas-based plants 322 generation adequacy 324–5 and Germany 326, 333–4 hydropower 322, 324, 326, 329 interconnection/interconnectors 322, 323, 325, 326, 333–4 internal market 333 intraday market 327 and Lithuania 326, 327 market characteristics 321–2 market liberalization 325–6 Ministry of Petroleum and Energy 323, 330–1, 333 NASDAQ OMX Commodities 327 and Netherlands 323, 326 non-renewable energy 326 Nord Pool 325, 326–7 Market Surveillance 331 Spot 326–7, 331 NVE (energy regulator) 323, 324, 327, 328–9 Energy and Regulation Department 330 options 327 peak demand 326, 332 Power Pool 325 pre-emption, right of 324 price areas 327 price hedging 327 quotas 329 red-green coalition government (2005–2013) 321, 332 regional grid 330 regulation power market 328 regulatory framework 322–4 renewable generation sources (RES) 322, 328–9, 332 retail trade 328 reversion doctrine 324 right-wing government (2013-current) 321–2, 332 risk management 327 and Russia 323, 326 sales licence 324 security of supply 325, 326, 332, 333, 334 social efficiency of cross-border participation from EU perspective 113n stability of supply 326 Statnett (transmission system operator (TSO)) 323, 325, 328, 330 strategic reserve 332 and Sweden 321, 323, 326–32 system price 327 thermal generation 322 Third Energy Package 331 transmission system operator (TSO) 327 see also Statnett and UK 326, 333–4

394

Index

Norway (cont.) uranium 322 vertical integration 323 Water Resources and Energy Directorate 322–3 wholesale trade 327 wind generation 322 nuclear generation 11, 158 objective justification test 17–18 Ockenfels, A. 9n off-switchable blocks 27 Office of Gas and Electricity Markets (OFGEM) 6, 8n, 26, 194, 198 oil-based power 6, 20 open tender procedure 165 operating costs 143 opportunity costs 43 opt out 107, 122 option contracts 14, 37, 69 option fee 133 option premium 14 outages see blackouts overcapacity 8, 25, 36, 81, 130 cross-border participation 99, 112, 114, 117 energy-only market sustainability 63 multilateral 51 renewable energy and demand participation 151 overinvestment 52 Panama 13 Parmigiani, L. 19n, 26n partial capacity mechanism 72 pay-as-clear auction 190–2 peak-load capacity/plants 140, 153 assessment of adequacy needs and cross-border resources 49 capacity payments and market efficiency 149–50 forward 100 optimal generation mix 142–7 price caps and investment level distortion 148 renewable energy and demand participation 151–2 peak-load pricing theory 34, 61–2, 80, 142 peaking unit construction costs 113 peaking units investment 115–16 penalties 13, 134–5, 139 penalty insurance market 197 performance incentives 8 performance tests 9 periodic capacity auctions 193 Peru: buying side of auction 122, 123 critical period 131 lag period (lead time) 128 Proinversión (Private investment Promotion Agency) auctions 127n–8n, 129 RER auctions 127n selling side of auction 126 Pfeifenberger, J.P. 4n photovoltaic/solar power 113, 117, 130, 158 physical and economic fundamentals of capacity rights trade 102–4 Physical Transmission Rights (PTRs) 100–1, 103n–4n

Pivotal Supplier Index (PSI) 194n Poignant, S. 257 Poland 231, 335–50 bilateral contracts 335 biomass 338 capacity auctions 13 capacity certificates 346 capacity charge 346 ‘capacity debate’ 338 capacity measures 342–4 capacity reserve margin 346 centralized capacity mechanism (vs UK model) 345–6, 347–8 Centrally Dispatched Generation Units 345 coal-fired power plants 335, 338–40, 348 cold intervention reserve 342–3 contracts for difference (CfDs) 345, 346–7, 348–9 and Czech Republic 336 day-ahead market 336 decentralized capacity mechanism (vs French model) 345, 346, 347 decommissioning 338, 340, 341, 342 Demand Side Response (DSR) contracts 340, 344, 348 emergency import 341n ENEA 336, 340, 341n ENERGA 336 Energy and Environmental Guidelines (EEAG) 2014–2020 347–9 Energy Regulatory Office (URE)(energy regulator) 337, 344 Investment Reports 337, 341–2 National Reports 337 energy three-pack 339n energy-only market 342–7 environmental standards 338 European dimension 347–9 Flow-Based Market Coupling project (FBMC) 336 forecast demand 341 and France 335 gas-fired plants 335, 340, 342, 348 generation adequacy 336–42 and Germany 335, 348 Grid Code 343, 348 and Hungary 336 interconnection/interconnectors 336, 348–9 KGHM 341n Kompania Weglowa 340 lignite-fired power plants 335, 338–40 limited lifetime derogation 338, 342 and Lithuania 348 market characteristics 335–6 market coupling 336 market-wide capacity mechanism 344–7, 348–50 Ministry of Economy 336–7, 339, 346, 347, 350 Report (2013) and follow-up actions 337, 340, 341, 344–5 North-Western European (NWE) market coupling 336 nuclear generation 341, 345, 346 operating capacity reserve 343–4, 348

Index peak demand 344, 346 PGE 336, 340, 341n, 343 PGNiG 339n POLPX power exchange 336 Power Generating Scheduling Units (JGWa) 343 power purchase agreements (PPAs) 349 Price Coupling of Regions 336 PSE 337, 340, 341, 343–4 public service obligation (PSO) 348 qualitative coefficient 344 refurbishment 341 regulatory framework 336–7 renewable generation sources (RES) 335, 338, 339n, 340, 346 intermittent 338 reserve capacity 348 and Romania 336 second-best option 346 security of supply 336–7 and Slovakia 336 state aid 346, 347, 348 strategic reserve 11 strike price 346–7 SwePol Link 336 TAURON 336, 339n, 340, 341n, 343, 348 temporary charge 349 Third Energy Package 339n timeline 347 transmission system operator (TSO) 335, 337, 340, 341–6, 350 and UK 335, 347, 348–50 wholesale price 338–9, 346–7 wind generation 341–2 policy options 67–72 design options and market failures 69–72 reform without capacity mechanism 67–8 portfolio effects 73 Portugal 15–16, 85, 91, 120, 213, 254 and Spain 352, 356 pre-qualification 196, 199 price caps 4, 87, 153 antitrust law 186, 197 capacity payments and market efficiency 149 coordination rationale and potential 80, 81 design options addressing market failures 70 energy-only market sustainability 67 and investment level distortion 147–9 non-compliance penalties 135 optimal generation mix 144 policy options 69 renewable energy and demand participation 151, 152–3 rights over system resources during extreme scarcity 52n selling side of auction 126 system adequacy problems 138 price controls 209 price spikes 34, 67, 111 price-based capacity mechanisms 10, 84 price-fixing 187 price-inelastic consumers 65 producer surplus 34n, 61 product market 187–8 production costs 143

395

production forecast inaccuracies 72 prohibition of customs duties and charges of equivalent effect on imports and exports 203–4 prohibition of discriminatory internal taxation 204–6 prohibition of quantitative restrictions and all measures having equivalent effect 206–10 proportionality 174, 184, 211–13 public policy and public security 214–18 public service obligation (PSO) 17–18, 23, 164–6, 176 ramp constraints 41n, 43 rationing 3–4, 54, 105–6, 131, 144–6, 149, 153 real-time pricing 66 reference market 132, 139 reference model 91 reform with capacity mechanism 77–8 reform without capacity mechanism 67–8 Regional Initiatives (RIs) 85 regulatory arrangements 79 regulatory failures 28–9, 80, 87, 173 reliability options 12, 14–15, 37, 38n, 154, 185–6, 188 contract 134 contract duration 129 coordination rationale and potential 81 cross-border participation 101 design options addressing market failures 71 policy options 69 reliability pricing model 71n reliability product requirements 130–6 in capacity- and energy-constrained systems 130–1 critical period 131–3 penalties for non-compliance 134–5 tradable quantity constraints 135–6 reliability as a public good 177 reliability rights 99–100, 103, 117 reliability standard 49 reliability of supply 96–7, 99, 106–7, 110, 116 remuneration of cross-border resources 51–2 remuneration, excessive 64–5 remuneration of flexibility and impact of capacity mechanism 41–4 renewable generation sources (RES) 5–6, 21, 32, 40–1, 140 balancing obligation 72–3 capacities 21, 113, 118 capacity payments 15–16 Commission Communication (November 2013) 24, 25, 26 critical period 133 and demand participation 150–3 deployment 30 Energy and Environmental Guidelines (EEAG) 2014–2020 7, 88, 171 extent of renewable support 74 free movement of goods 205, 207–9, 219–22 intermittent 41–3, 45, 47, 49, 55 and market design 72–5 production 51 remuneration of flexibility 41 state aid rules and funding 158–9, 167, 176

396

Index

renewable generation sources (RES) (cont.) subsidized 79 see also hydropower; photovoltaic/solar power; thermal generation; 20/20/20 Package; wind generation Renewable Generation Sources (RES) Directive (2009) 21 reserve capacity 6, 18, 60, 70 reserve margin 45–6, 55, 100, 106, 117 Residual Supply Index (RSI) 194n retail price regulation 28–9 risk allocation 51–2 risk-hedging 14, 37 rolling blackouts see rationing Romania 229, 336 Roques, F. A. 4n Russia 88n, 204, 323, 326 Sadowska, M. 364 Savva, N. S. 4n scarcity conditions/situations: buying side of auctio n123 coordination rationale and potential 80 critical period 132–3 cross-border participation 98, 112–15 energy-only market sustainability 67 explicit cross-border participation with congestio n106 non-compliance penalties 134 policy options 68 reliability in capacity- and energy-constrained systems 130, 131 selling side of target market 126 system adequacy problems 138, 139 see also blackouts; brownouts scarcity insurance premium 55, 62 scarcity price bidding 8 scarcity prices 34, 61, 65, 69–70, 81 scarcity rents 34n, 116 Schoser, C. 254n screening curves 143 secondary certificate markets 185 security of supply 6, 36, 38–9 coordination rationale and potential 81, 85, 89 energy-only market sustainability 60 as ‘public good’ 65–6 renewable energies and market design 72 self-sufficient capacity mechanism, costs of 44–8 self-supply 13 selling side 125–7, 138 Serbia 229 service of general economic interest (SGEI) 23–4, 162–3, 165–6, 176 service of general interest (SGI) 166 settlement prices 68 short run marginal cost (SRMC) 86, 133 short-term price of reference market 139 shortage event 134–5 see also scarcity conditions/situations Shortage Event Availability Score 135 Sido, B. 257 Slovakia 6, 336 Slovenia 229 smart meters 9, 66, 123n, 145n, 152

social efficiency: from EU perspective 110–17 from national perspective 106–10 gains 117 soft caps 134 solar power see photovoltaic/solar power solidarity principle 52–3, 55, 105–6 South America: buying side of auction 124 contract duration 129 indexation formulas 137 lag period (lead time) 127 selling side of auction 126 system adequacy problems 120–1, 138 see also Argentina; Brazil; Chile; Colombia; Peru Spain 254, 351–64 administrative certificate 360 Administrative Register of Electricity Production Facilities 359 Agency for Cooperation of Energy Regulators (ACER) 356 and Austria 364 authorization procedures for new generation capacity construction 355 Autonomous Community 355 availability service 357–9 biomass 351 blackouts 353 capacity payments 15–16, 356–9, 364 capacity price curve 358 coal-fired plants 351–2, 354, 359 cogneration units 351, 355 combined cycle gas turbine (CCGT) power plants 44, 354, 359 compensation 363–4 competition distortion 361 congestion 353 coordination rationale and potential 81–2, 84–5, 89, 91 decommissioning 354 and Denmark 364 economic crisis 353, 358, 362, 364 efficiency 358 electricity deficit 351 electricity pool 352 Electricity Sector Act (2013) 354, 362–3 electricity tariff deficit 354–5 Endesa 352 Enel 352 energy market design 59 E.ON 352 European dimension 363–4 explicit cross-border participation with congestion on interconnections 102n FADE 354 first come, first served rule 355 flexibility 353 and France 352 Gas Natural Fenosa 352 gas-fired plants 351–2, 353, 354, 358 generation adequacy 353–4 and Greece 356 grid planning 353 HC Energia 352

Index hydropower 351, 354, 359, 362 Iberdrola 352 interconnection/interconnectors 352 investment incentives 357–8 and Ireland 356 and Italy 356 judicial review of regulations 360–2 legitimate expectations, principle of 361–2 marginal pricing pool 352 market characteristics 351–2 market liberalization 351–2, 354, 356 Ministerial Order 356–7, 358, 360, 363 Ministry for Industry, Energy and Tourism 353, 360, 363 and Morocco 352 National Commission for the Markets and Competition (CNMC) 353, 354, 358, 363 non-discrimination 355, 357, 361, 362 nuclear power 351, 360, 362 objectivity principle 355, 361 oil-fired plants 358, 359 overcapacity 353 payments for investments 358 peak demand 353 photovoltaic/solar power 351 and Portugal 352, 356 public service obligation (PSO) 364 reduction and abolishment of some capacity payments 2012–13 359–60 REE (transmission system operator (TSO)) 353, 358 Operating Procedure 355 regulatory failure 354 regulatory framework 352–3 renewable generation sources (RES) 352, 353, 354, 355, 358, 363 intermittent 351 reserve margin 358 Royal Decree 355–6, 359, 360 security of supply 353–4, 356, 358, 361, 363, 364 self-sufficiency and excess capacity 45 Service of General Economic Interest (SGEI) 364 spot market 352 state aid rules 361, 363–4 stranded costs 356–7, 363–4 subsidies 354 system adequacy problems 119 tariffs and charges, distinction between 362–3 thermal generation 351, 353–4, 362 transmission adequacy 353 transmission bottlenecks 357 transparency 355, 357, 358, 361 UNESA 361–2 wind generation 351–2 Spees, K. 4n spill-over effects 36 spot market 125, 142, 153, 195 bidding 185 prices 1, 17, 69, 131, 138, 145 stability of policy framework 68 standstill provision 181 state aid 86–8, 203

397

state aid control and ‘missing money’ problem 157–81 assessment 175–80 Energy and Environmental Guidelines (EEAG) 2014–2020 171–9, 181 policy evolution 158–62 prohibition of customs duties and charges of equivalent effect on imports and exports 179–80 prohibition of discriminatory internal taxation 179–80 State Aid Modernization (SAM) 161–2 see also funding capacity mechanisms and state aid rules State Aid Modernization (SAM) 161–2, 167, 175, 177 state resources financing 166–70 Stoft, S. 9n, 145n storage 173, 174 stranded costs and compensation 74, 159–60, 171 strategic bidding 189 strategic reserve 10–12, 37, 69, 70, 81, 185–6, 188 stress events 45, 49, 51, 133, 178 see also scarcity conditions/situations strike price 17, 37, 38n, 186 critical period 132–3 cross-border participation 101n policy options 69 reliability options 14–15 system adequacy problems 138 subsidiarity principle 97 subsidies 63, 149, 168 substantive tests 179 suitability test 184 sunk capital costs 151 sunk fixed costs 126 supply of capacity 7–8, 81 Sweden 254 antitrust law 184–5 electricity certificates market 328–9, 330 Energy Agency 209, 328–9 free movement of goods 208, 221–2 and Norway 321, 323, 326, 327, 328–9, 330–1, 332 peak-load reserve mechanism 370–1 strategic reserve 11 Switzerland 45, 89, 113n, 229, 254, 312–13 and Germany 276n, 278, 286 and Italy 303, 312–13 system adequacy problem 119–39 buying side 121–5 centralization 123–5 contract duration 128–30 design elements 121 existing plants 125–6 financial warranties 137–8 indexation formulas 137 lag period (lead time) 127–8 recommendations 138–9 selling side of auction 125–7 target market 121–7 technological neutrality 127 see also reliability product requirements system resources, rights over, during times of extreme scarcity 52–4

398

Index

target market 121–7 Target Model 83–4, 104, 291, 299 targeted capacity mechanisms 10 technological neutrality 30, 127, 150–4, 174 technology-specific tenders 138 tendering 15–16, 19–20, 138, 187 THEMA consulting group 45 thermal generation 6, 129–31, 136 Third Energy Package (July 2009) 83 threshold price 35, 37 time frames 26, 38, 41, 175 Tirole, J. 141n, 145n Tol, R. S. J. 144n total cost 149–50 total production cost 146–8 tradable quantity constraints 135–6 transaction cost 71 transmission system operator (TSO) 17, 37–8 anticompetitive practices 185, 187, 199–200 capacity auctions 13 Commission Communication (November 2013) 23, 25, 27, 29 coordination rationale and potential 84, 89–90, 93, 94 cross-border participation 97–103 demand side response (DSR) 9n energy market design 59 explicit cross-border participation with congestion on interconnections 106 reliability options 14 remuneration of flexibility 43 renewable energies and market design 72, 75 rights over system resources during extreme scarcity 53–4 risk allocation and remuneration of cross-border resources 52 social efficiency of cross-border participation from EU perspective 113 social efficiency of cross-border participation from national perspective 107, 109–10 state aid rules and funding 160n, 163n, 168, 170, 178 strategic reserve 10 supply and demand of capacity 7 see also European Network of Transmission System Operators for Electricity (ENTSO-E) transparency 39, 47, 88, 138 Turkey 289 Tusk, D. 340n 20/20/20 Package 18–22, 159 tying 27 underinvestment due to prohibitive price risk (uncertainty for investors) 66–7 United Kingdom 6, 246n, 254–5, 317, 365–82 anti-bribery rules 379 anticompetitive practices 182, 185, 190, 194, 197–8 assessment of adequacy needs and cross-border resources 51 assessment of options 370–1 auctions 13, 374, 376–7 balancing arrangements 365, 370, 381 Balancing and Settlement code 378

barriers to entry 367 bilateral contract 365 buying side of target market 124 capacity demand curve 375 capacity gap 366–9 capacity margins 370 capacity market design, refinement of 372–3 capacity market options, assessment of 371–2 Capacity Market Rules 376 capacity market selection as preferred option 372 Capacity Market Settlement Agent (Elexon) 378 capacity market structure 373–8 Capacity Market Warnings 377 capacity mechanism 365 capacity payments 366 capacity withholding 379 Carbon Price Floor scheme (2013) 179 carbon price support 366 cash-out price 367, 370, 371 Certificate of Ethical Conduct 379 coal-fired plants 365 collusion 379 combined cycle gas turbine (CCGT) power plants 374 Commission Communication (November 2013) 26, 27n, 29 competition rules 379 coordination rationale and potential 81–2, 84–5, 89 Cost of New Entry (CONE) 374 net 375, 377 critical period 133 de-rating 367–9, 380–1 delivery 377–8 demand curve parameters 375 demand side response (DSR) 372, 375–7, 378, 379–80, 382 Department of Energy & Climate Change (DECC) 50, 52, 366, 370–1, 380–1 Capacity Market: Design and Implementation Update (2012) 368 Capacity Market Gaming and Consistency Assessment Report 379 Capacity Market Strawman 372–3 de-rated capacity margin estimates 369 draft Electricity Market Reform Delivery Plan (2013) 374–5 Electricity Market Reform: Capacity Market Detailed Design Proposals (2013) 372 Electricity Market Reform: Consultation on Proposals for Implementation (2013) 373 Electricity Market Reform (2010) 366, 368, 370–1 Electricity Market Reform Delivery Plan (2013) 375 Impact Assessments (2012-14) 368–9 Planning our electric future White Paper (2011) 368, 371, 372, 380 Statutory Security of Supply Report (2013) 369 descending clock basis 377 determination of capacity required 374–5 dispute resolution process 376

Index Dynamic Dispatch Model 369 Electricity Market Reform (EMR) 177, 366, 375, 380 see also under Department of Energy & Climate Change (DECC) eligibility 375–6 emissions performance standard 366 Energy and Environmental Guidelines (EEAG) 2014–2020 175 energy market design 59 enforcement mechanism 378–9 Environmental and Energy State Aid Guidelines 380 European dimension 380–1 exclusion of interconnected capacity 380–1 exiting the capacity market 378 expected energy unserved (EEU) 367–8 feed-in tariffs 366 free movement of goods 212–13, 214 gaming 379 gas-fired power 365 generation adequacy 22, 366–70 imports 365 insider trading 380 interconnection/interconnectors 378 legislative framework 373 liquidity 378 load increasing 379 load-following 377 Loss-of-Load Expectation (LOLE) 374, 375 market characteristics 365 market distortion 371, 379–80 market failures 366–7 market manipulation 380 market-wide mechanism 372 minimization of necessity for capacity mechanism: current market improvement 369–70 ‘missing money’ problem 367, 368, 370, 381 National Grid (transmission system operator (TSO)) 52, 178, 365, 371n, 377, 381 Electricity Capacity Report (2014) 375 and Norway 326, 333–4 nuclear power 365, 368 Office of Gas and Electricity Markets (OFGEM) 367n, 375, 376, 378, 380 Electricity Capacity Assessments 368–9 estimates of de-rated capacity margins 369 Project Discovery 366 Statutory Security of Supply Report (2013) 369 one-year capacity agreement 377 optimal generation mix 144n over-the-counter transactions 365 overcompensation 381 pay-as-clear basis 376 payment 378 peak demand 378 and Poland 335, 347, 348–50 pre-qualification criteria 381 price caps 367, 375 price makers 377 price takers 377, 381 regulatory framework 366

399

reliability in capacity- and energy-constrained systems 131 reliability options 15 reliability standard 367, 374, 375 remuneration of flexibility 41 renewable generation sources (RES) 365, 370 Scottish Power 365 secondary market 377 security of supply 368–9 selling side of auction 126n Settlement Body 168 settlement periods 365 short-run marginal cost 378 Short-Term Operating Reserve (STOR) mechanism 371 Single Electricity Market 365n slippery slope effect 371 social efficiency of cross-border participation from EU perspective 113n spot markets 365 SSE 365 state aid rules and funding 158, 170, 176n, 178–80, 181, 380–1 storage 372 strategic reserve 371–2 stress periods 368, 377, 381 system adequacy problems 120 target capacity 371, 375 Target Model 381 tender for targeted resource (TTR) 370–1 tenders for capacity 366 three-year capacity agreement 377 transmission system operator (TSO) 367, 377, 378 see also National Grid value of lost load (VOLL) 368, 374, 375, 378 wait and see approach 367 wholesale market 365, 367 wind generation 368 United States 15, 19n, 43, 59, 128, 370 antitrust law 189 Federal Energy Regulatory Commission (FERC) 380 Forward Capacity Market 120 Forward Capacity Market of Independent System Operator New England (ISO-NE) 14, 122, 124, 126, 128, 129, 134–5, 136 and France 259 James Bay hydro plant (power trade from Quebec to Massachusetts) 103 MISO 105n New York Independent System Operator (ISO) 14 Pennsylvania-New Jersey-Maryland Interconnection (PJM) Independent System Operator (ISO) 14, 71n, 105n, 119–20, 122, 124, 128 Reliability Pricing Model 120, 122 use-it-or-lose-it rule 104 value of lost load (VOLL) 4, 34, 37, 80, 142, 144–5, 148 variable costs 61–2, 143, 146 Vázquez, C. 134 verification 89

400 vertical integration 124–5, 138, 183, 185 volatility of generation 72, 74–5 volume-based capacity mechanisms 10 wait-for-the-tender approach 13 ‘waterbed effect’ 70 welfare distribution effects 36 wholesale market 186, 188, 189 clearing 61 interconnectedness 33 wholesale prices 62–3, 70

Index caps 88, 173 regulation 28–9 Willems, B. 364 wind generation 117, 150–1, 153–4, 158, 169, 180 reliability in capacity- and energy-constrained systems 130 tradable quantities constraints 136 withholding capacity 185, 191–2 zero profit 146–7 Zöttl, G. 147n