Reforming the Chinese Electricity Supply Sector: Lessons from Global Experience [1st ed.] 9783030394615, 9783030394622

Citation preview

Michael G. Pollitt

Reforming the Chinese Electricity Supply Sector Lessons from Global Experience

Reforming the Chinese Electricity Supply Sector

Michael G. Pollitt

Reforming the Chinese Electricity Supply Sector Lessons from Global Experience

Michael G. Pollitt Judge Business School University of Cambridge Cambridge, UK

ISBN 978-3-030-39461-5    ISBN 978-3-030-39462-2 (eBook) https://doi.org/10.1007/978-3-030-39462-2 © The Editor(s) (if applicable) and The Author(s), under exclusive licence to Springer Nature Switzerland AG 2020 This work is subject to copyright. All rights are solely and exclusively licensed by the Publisher, whether the whole or part of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission or information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed. The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use. The publisher, the authors and the editors are safe to assume that the advice and information in this book are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or the editors give a warranty, expressed or implied, with respect to the material contained herein or for any errors or omissions that may have been made. The publisher remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. This Palgrave Macmillan imprint is published by the registered company Springer Nature Switzerland AG. The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland

To Christian Romig, formerly of the British Embassy in Beijing, for inspiring my interest in Chinese power market reform. Without Christian’s enthusiasm, energy and charm this book would never have been written.

Preface

The Chinese electricity sector is the largest in the world, covering over 25% of world electricity supply and responsible for perhaps 9% of global greenhouse gas emissions (by 2016). China’s electricity companies such as the State Grid Company of China and China Southern Grid are among the world’s largest companies. Many other countries, including the UK, liberalised their electricity systems in the 1990s, creating competitive wholesale and retail electricity markets and separate incentive-regulated electricity grids. This has resulted in electricity systems very different from the pre-liberalisation era. China has only slowly moved towards reforming its electricity sector, but in recent years it has begun an ambitious new round of reforms aimed at introducing competitive wholesale electricity markets and incentive regulation for its power grids. This was initiated in March 2015 by the No. 9 document on ‘Deepening Reform of the Power Sector’ published by the China State Council with the aim of offering lower prices to China’s industrial electricity customers. This book seeks to provide lessons for China’s reforms from international

vii

viii Preface

experience, combining a detailed review of reform lessons from around the world, a specific application to China and a particular focus on how exactly the industrial price of electricity is determined in a liberalised power system. The book draws on the outputs of a three-year research programme based on the interactions of Chinese and British power market professionals, facilitated by the British Embassy in Beijing. Cambridge, UK

Michael G. Pollitt

Acknowledgements

I owe many debts of gratitude for the writing of this book. The British Embassy in Beijing has supported this project consistently since October 2015 when I was invited to participate in the UK-China Energy Dialogue around the time of the visit of President Xi to London. Since then the Embassy, together with British Consulates in Guangzhou and Shanghai, has hosted me on eight visits to China and introduced me to a wide range of different stakeholders from across the Chinese electricity supply sector. I particularly wish to thank Christian Romig, then of the British Embassy in Beijing, to whom this book is dedicated. It was through Christian that I have been inspired to spend four years (and counting) working on Chinese power market reform. It was as Christian and I travelled in an Embassy car between meetings in Beijing that I felt personally challenged—by the Lord himself (?)—to take Chinese power market reform seriously! I am a huge admirer of the dedicated staff of the UK Foreign Office in China and would wish to thank the many officials who worked to make my visits to China such enjoyable and productive experiences. The British Embassy in Beijing also generously arranged for the translation of the academic paper versions of Chaps. 2, 3 and 4 into Chinese and co-sponsored a joint conference with the University of Cambridge around Chap. 2 in May 2017.

ix

x Acknowledgements

I readily acknowledge my co-authors on the individual chapters: Chung-Han Yang, Hao Chen and Lewis Dale. Chung-Han and HaoChen helped me access the many Chinese papers on power market reform and provided me with numerous insights into the way that the reforms are viewed in China, as well as providing superb support for the writing of Chaps. 2 and 3. Chung-Han was a very agreeable companion on my first trip to Guangdong. Both were exemplary members of the Energy Policy Research Group (EPRG) during their time in Cambridge. Lewis Dale has supported this project from the start, travelling to China twice and providing me with first-hand insights into the UK’s own electricity reform. Lewis has also answered many of the questions we have received over the course of the project about power market reform from Chinese stakeholders and was always willing to meet Chinese power sector visitors to the UK. Lewis is a co-author of Chap. 4. In addition, I would wish to thank the many Chinese stakeholders who have met me over the last five years to talk about power market reform. The topic is huge and without them I would not have been able to begin to get a sense of what the issues were for China. I would wish to thank all those stakeholders I have met in Beijing, Guangdong, Yunnan, Fujian, Zhejiang and Jiangsu for their time and generosity in answering my questions and explaining their insights to me. Thanks go to Professor Yongsheng Feng of the Chinese Academy of Social Sciences. Yongsheng oversaw the publication of academic paper versions of Chaps. 2, 3 and 4 in one of the Academy’s journals, Financial Minds, providing many judicious editing comments and much encouragement on the papers in the process. I would also wish to thank Rachel Chen, Bai-Chen Xie, Jun Xu and Geoffroy Dolphin for their help with the editing of the book manuscript. Finally, I should acknowledge all of the support I have received from my own university. The book is an output of the In Search of ‘Good’ Energy Policy grand challenge project of Energy@Cambridge (which has now become Energy Transitions@Cambridge). I would wish to thank Professor David Newbery of the EPRG, Dr Isabelle de Wouters and Dr

 Acknowledgements 

xi

Shafiq Ahmed of Energy@Cambridge in particular for their support for this work. I am happy to acknowledge funding from the ESRC Global Challenges Research Fund (which supported the work behind Chap. 2), an ESRC Impact Acceleration Award (for Chap. 3) and funding from the Judge Business School impact fund (for Chap. 4). Finally, I also wish to thank my family—Yvonne, Daisy and Sammy—for all of their support and willingness to put up with my trips to China!

Contents

1 An Introduction to the Chinese Power System and Its Reform  1 1.1 Background to Our Study   1 1.2 The Structure of the Book   2 1.3 The Scale and Scope of the Chinese Electricity Supply Industry in 2015   4 1.4 The Structure and Organisation of the Chinese Power Sector 10 1.5 The History of Power Sector Reform in China  13 1.6 What Motivated the 2015 Power Sector Reform?  16 References 19 2 Lessons for China from International Experience of Power Sector Reform 23 2.1 Introduction  23 2.2 Market Restructuring and Ownership Changes  26 2.2.1 Vertical Separation (1) and Horizontal Restructuring (2)  26 2.2.2 The Creation of Wide Area Independent System Operators (3)  34 2.2.3 Privatisation and Monopolies (4)  40 xiii

xiv Contents

2.3 Supportive Secondary Market Arrangements  44 2.3.1 Creation of Spot and Ancillary Services Markets to Support Real-Time Balancing of the System (5)  44 2.3.2 Participation of Demand Side in Wholesale Electricity Markets (6)  47 2.3.3 Regulated Third-Party Access to, and Efficient Allocation of, Scarce Transmission Capacity (7)  49 2.4 Appropriate Economic Regulation  55 2.4.1 Unbundling of Regulated Network Charges and Competitive Segment Charges (8)  55 2.4.2 Mechanisms to Ensure Competitive Procurement of Wholesale Power for Regulated Final Customer Groups (9)  58 2.4.3 The Creation of Independent Regulatory Agencies to Regulate Monopoly Network Charges and Monitor Competitive Segments (10)  60 2.5 Efficient Promotion of Low Emission Technologies  69 2.5.1 Competitive Procurement Processes for Low Carbon Generation, with Some Exposure to Wholesale Price Variability (11)  69 2.5.2 Cost Reflective Access Terms for Renewables (12)  74 2.5.3 Appropriate Pricing of Environmental Externalities (Both Carbon Dioxide and Other Atmospheric Pollutants, such as Sulphur Dioxide) (13)  76 2.6 All Good Power Market Reforms (and Indeed, Significant Economic Reforms More Generally) Involve Appropriate Transition Mechanisms (14)  80 2.6.1 Theoretical Significance  80 2.6.2 General Reform Experience  80 2.6.3 Chinese Context  81 2.7 Conclusions  85 2.7.1 International Lessons and Policy Priorities for China 85 2.7.2 Suggestions for Future Research  88 Appendices 90

 Contents 

xv

Appendix I: Calculation Process of the Switching Carbon Prices from Coal- to Gas-Fired Power Plant Investment in 2015 (LCOE=levelised cost of electricity)   90 Appendix II: Calculation Process of the Switching Carbon Prices from Coal- to Gas-Fired Power Generation in 2015   91 References 92 3 Power Market Reform in China: Lessons from Guangdong103 3.1 Introduction 103 3.2 Background 106 3.2.1 Guangdong Within China 106 3.2.2 The Size of the Electricity Sector in Guangdong 109 3.3 How the Power Market Works 113 3.3.1 International Context 113 3.3.2 The Power Market in Guangdong 116 3.4 New Players 119 3.4.1 International Context 119 3.4.2 New Energy Market Players in Guangdong 122 3.5 Effects on Operations and Dispatch 126 3.5.1 International Experience 126 3.5.2 Effects on Dispatch in Guangdong 128 3.6 Key Points for Improvement 130 3.6.1 Discussion of Overall Impressions of Reform 130 3.6.2 Recommendations for Furthering Reform 139 Appendix: How Changing Infra-Marginal Bids Changes the Auction Results in the 2016 Power Market Auction Design  141 References145 4 How Industrial Electricity Prices Are Determined in a Reformed Power Market: Lessons from Great Britain for China153 4.1 Introduction 153 4.2 How Is the Industrial Electricity Price Set in Great Britain154 4.3 The Key Actors in the Electricity System in Great Britain 156 4.4 Wholesale Prices 161

xvi Contents

4.5 Retail Margins 169 4.6 Regulated Network Charges Determination 173 4.7 Transmission Charges 182 4.8 System Balancing Charges 190 4.9 Distribution Charges 201 4.10 Environmental Levies and Taxes 204 4.11 Overall Lessons on Price Determination for China from Great Britain 207 References209 5 Prospects for Reform of China’s Electric Power Sector215 5.1 High-Level Messages from Previous Chapters 216 5.2 Recent Developments on Power Sector Reform in China 217 5.3 Suggestions for Next Steps 224 5.3.1 Improving Regulatory Capacity 225 5.3.2 Improving Regulatory Reporting 225 5.3.3 Promoting Learning from the Pilot Markets 226 5.3.4 Putting All Generation and Demand in the Wholesale Market 227 5.3.5 Consider Whole or Part Privatisation of One Large Generator 228 5.3.6 The Creation of Genuine Interprovincial Market Should Be Done in Stages 228 5.3.7 Pay Attention to Mitigation of the Social Effects of Power Sector Reform 229 5.4 Fundamental Questions Raised by China’s Power Market Reform230 5.4.1 Is China Ready for the Full Implications of Electricity Markets? 231 5.4.2 Is There an Easier Way to Deliver the Benefits of Power Market Reform Than the Route Currently Being Pursued? 231 5.4.3 Can Reform Be Sustained and Completed in China Given Its Institutional Set-Up? 232

 Contents 

xvii

5.4.4 Can China Have a Successful Power Market Reform Without Widespread Private Ownership of the Sector? 233 5.4.5 Is China Willing to Break Up SGCC and CSG in Ways That Will Promote Power Market Development?234 5.4.6 How Will China Combine Power Sector Reform with Decarbonisation? 234 5.4.7 What Will Reform Reveal About Chinese Electricity Consumers? 235 5.5 Key Closing on Messages for Chinese Electricity Stakeholders on How to Approach Power Sector Reform 236 5.5.1 Policy-makers 236 5.5.2 Regulators 237 5.5.3 Generators 237 5.5.4 Retailers 237 5.5.5 Grid Companies 238 References238 Index241

About the Authors

Hao Chen  is an energy economist based in China. He recently graduated with a PhD in Energy Economics from the Beijing Institute of Technology. He is currently  Associate Professor of Economics at the China University of Geosciences in Wuhan. Lewis  Dale is Regulation and Strategy Manager at National Grid Electricity Transmission, UK.  Dale has a career spanning more than 30  years in the UK electricity sector, witnessing electricity reform at first hand. Michael  G.  Pollitt is Professor of Business Economics at the Judge Business School, University of Cambridge, UK. He is an Assistant Director of the Energy Policy Research Group (EPRG) and a joint Academic Director of the Centre on Regulation in Europe (CERRE). Pollitt is on the council of the International Association for Energy Economics (IAEE). He is one of Europe’s leading energy economists and has written extensively on the impacts of electricity reforms across the world. Chung-Han Yang  is an environmental and energy lawyer from Taiwan. He recently graduated from the University of Cambridge with a PhD in Environmental Law. He is a partner of Dentons LLP and an adjunct assistant professor at National Tsing Hua University, Taiwan. xix

List of Figures

Fig. 1.1 Fig. 1.2 Fig. 1.3

Fig. 1.4

Fig. 1.5

Fig. 1.6 Fig. 2.1 Fig. 2.2

The size of the Chinese electricity sector. (Source: NBS (2016)) 5 The scale of power plant capacity in construction. (Source: China Electricity Council (2015)) 6 China’s electricity consumption slowing down? (Source: CEC 2013 website, Available at: http://www.cec.org.cn/ guihuayutongji/tongjxinxi/niandushuju/2013-04-19/100589.html)7 China’s global share in the fuel consumption for its power generation. (Source: BP’s the Statistical Review of World Energy 2016, Available at: http://www.bp.com/en/global/ corporate/energy-economics/statistical-review-of-worldenergy/downloads.html)8 China’s global share in CO2, coal and electricity production. (Source: BP’s the Statistical Review of World Energy 2016, Available at: http://www.bp.com/en/global/corporate/energy-economics/statistical-review-of-world-energy/ downloads.html)9 Reform timeline for electricity sector. (Source: An et al. (2015, p. 6)) 13 Relative performance of State Grid and China Southern Grid. (Sources: SGCC (2015) and CSPG (2015)) 30 Regional and provincial grid control areas in China. (Source: Derived from Wang and Chen (2012, p. 144)) 37 xxi

xxii 

Fig. 2.3 Fig. 2.4 Fig. 2.5 Fig. 2.6

Fig. 2.7 Fig. 2.8

Fig. 3.1 Fig. 3.2 Fig. 3.3 Fig. 3.4

List of Figures

Share of Big five generators in total capacity and total generation. (Source: Annual reports of power generation companies (2015)) 43 List of electricity product markets, including ancillary services products. (Source: Stoft (2002, pp. 82, 236)) 45 Sources of electricity demand in China and the US. (Source: NBS (2015) and EIA (2015)) 50 Current structure of regulatory bodies overseeing the Chinese electricity sector. (Notes: Ministry of Environmental Protection [MEP], National Development and Reform Council [NDRC], State Administration of Work Safety [SAWS], National Energy Association [NEA], State-owned Assets Supervision and Administration Commission [SASAC]. Source: Tan and Zhao (2016)) 64 Anti-monopoly institutions in China in 2016. (Source: Slaughter and May (2016, p. 2)) 65 International comparison of impact of reform on labour productivity in electricity distribution and transmission. (Notes: The six companies from Peru in this figure are Electro Sur Medio, Electrolima, Edelnor, Luz del Sur, Ede Chancay and Ede Cañete. CSG = China Southern Grid; SGCC = State Grid Corporation of China. Sources: Anaya (2010), China Electricity Council (2015), Domah and Pollitt (2001), Mota (2003), National Grid Electricity Transmission Report and Accounts, Pollitt (2004 & 2008)) 68 Map of Guangdong. (Source: https://wikitravel.org/upload/ shared/7/7a/Guangdong2.png)106 Timelines for reform in the Guangdong electricity sector. (Source: Adapted from An Bo et al. (2015, p. 6). Revised from Guangzhou Power Exchange (2017b)) 109 Generation capacity in the Guangdong electricity sector. (Source: Guangzhou Power Exchange Centre 2017a) 110 Electricity demand in Guangdong. (Source: (1) Data of the period from 2006 to 2016 are drawn from National Bureau of Statistics of China, http://www.stats.gov.cn/. (2) Data of 2017 are drawn from Guangdong power exchange centre: https://pm.gd.csg.cn)111

  List of Figures 

Fig. 3.5 Fig. 3.6 Fig. 3.7 Fig. 3.8

Fig. 3.9

Fig. 3.10 Fig. 3.11 Fig. 3.12 Fig. 3.13 Fig. 3.14 Fig. 3.15 Fig. 3.16 Fig. 3.17 Fig. 4.1 Fig. 4.2 Fig. 4.3 Fig. 4.4 Fig. 4.5

xxiii

Quality of service in Guangdong. (Source: Compilation of power industry statistics of China (2011–2017)) 112 Sources of electricity demand. (Source: Guangdong Power Exchange Center) 113 Power market prices (RMB/kWh). (Source: Guangdong Power Exchange Center, https://pm.gd.csg.cn) 117 Generator and retailer market shares in Guangdong in 2017. A. Generator capacity shares (Source: http://mp. weixin.qq.com/s/CBdmpsVFppV1j2WskWnPNQ). B. Retailer market shares. (Source: http://www.sohu. com/a/213958006_679911)123 Illustrative market clearing and price determination. A. Retailer (green) bids, generator (blue) offers and maximum trading volume (red). B. Calculation of system discount charge. C. Allocation of system discount charge to winning bidders. D. Calculation of final prices paid to winning retailers and generators. (Source: Jing et al. 2018) 134 Retailer (green) bids and generator (blue) offers, with maximum trading volume (red) 141 Calculation of system discount charge 142 Allocation of system discount charge to winning bidders 142 Calculation of final prices paid to winning retailers and generators143 Retailer (green) bids and generator (blue) offers, with maximum trading volume (red) 143 Calculation of system discount charge 144 Allocation of system discount charge to winning bidders 144 Calculation of final prices paid to winning retailers and generators145 The structure of the electricity industry in the UK 157 Wholesale market shares 2017. (Source: Ofgem (2018a), State of the Energy Market Report 2018, p. 50) 158 Business retail market shares 2017. (Source: Large-scale electricity profile class 5–8 +HH: Ofgem (2018a), State of the Energy Market Report 2018, p. 38) 158 The impact of X factors on the revenue of a regulated firm 177 NGET controllable operating costs. (Source: National Grid) 183

xxiv 

List of Figures

Fig. 4.6

Transmission investment. (Source: National Grid. Pre-1991 figures are based on CEGB information sources and are adjusted to reflect one-off investment in the GB-France interconnector)183 Loss of supply incidents at NGET. Note: Custom connections are for certain industrial customers who accept higher levels of interruption in return for lower tariffs. (Source: National Grid) 184 Energy not supplied by NGET. (Source: National Grid) 184 TNUoS 2018/19 regulated revenue. (Source: National Grid) 185 NGET’s 2008/09 reliability incentive scheme. (Source: National Grid) 187 The place of the balancing mechanism relative to real time. (Source: National Grid) 192 Balancing market information flows. (Source: Information from National Grid) 193 Bids in the balancing mechanism. (Source: National Grid) 194 Ranked bids in the balancing mechanism. BSAA = Balance Service Adjustment Action. (Source: National Grid) 195 National Grid transmission system operator constraint payments. (Source: National Grid) 196 National Grid electricity system operation external costs. (Source: National Grid) 197 Prices for firm frequency response. (Source: GB FFR Market Summary Reports from Aurora Energy Research https://www.auroraer.com)199 The development of real distribution revenue since privatisation average DNO in England and Wales. (Source: National Grid) 201 Improvement in quality of service since 1990. (Source: Ofgem)202

Fig. 4.7

Fig. 4.8 Fig. 4.9 Fig. 4.10 Fig. 4.11 Fig. 4.12 Fig. 4.13 Fig. 4.14 Fig. 4.15 Fig. 4.16 Fig. 4.17 Fig. 4.18 Fig. 4.19

List of Tables

Table 1.1 Table 1.2 Table 1.3 Table 1.4 Table 2.1 Table 2.2 Table 2.3 Table 2.4 Table 2.5 Table 2.6 Table 2.7 Table 3.1 Table 4.1 Table 4.2

Largest generators by capacity in 2014 Reform timeline for electricity sector in China Document No.9 of March 2015 and stated reform process Electricity price and fuel input price differential with US in 2014 Benchmark generation tariffs of coal-fired power plants (with FGD) in 2014 Demand participation in liberalised markets (DSR = demand side reduction) The potential for demand response in Shanghai and Jiangsu in selected years Expansion of Chinese transmission and distribution grid Evidence on civil service pay in China relative to stateowned companies Target share of non-hydro renewable energy in total electricity consumption in 2020 Electricity market pilot projects in 2016 Electricity price and fuel input price differential with US Breakdown of industrial price in the UK Menu regulation

11 14 17 18 33 48 51 54 66 72 83 112 155 178

xxv

xxvi 

List of Tables

Table 4.3 Table 4.4 Table 5.1 Table 5.2 Table 5.3 Table 5.4

Quality of service incentive scheme System operator services Characteristics of the pilot provinces in 2017 Provincial electricity balances in pilot areas in 2017 Electricity production mix in pilot provinces in 2017 Progress with power market implementation

180 191 220 221 221 222

1 An Introduction to the Chinese Power System and Its Reform

1.1 Background to Our Study This chapter introduces the Chinese power system and the background to the reform process which began in March 2015 with the publication of the No. 9 document of the China State Council on ‘Deepening Reform of the Power Sector’.1 Inter alia, this set of reforms seeks to introduce wholesale power markets to trade electricity under longer- and medium-­ term contracts and via spot markets. It also introduces incentive-based regulation of electricity network tariffs. In doing so it moves the pricing of both wholesale and retail electricity away from a regulated system of tariffs towards market-determined prices for industrial electricity customers. This represents a radical break with the past for China and a significant extension of the role of a market in a sector still dominated by massive state-owned utilities. The idea behind this book is that there is a wealth of global experience of power market reform that China can draw on in embarking on a renewed push to introduce competitive wholesale power markets and incentive regulation of power networks. The focus of the book is very  See China State Council (2015).

1

© The Author(s) 2020 M. G. Pollitt, Reforming the Chinese Electricity Supply Sector, https://doi.org/10.1007/978-3-030-39462-2_1

1

2 

M. G. Pollitt

much on the lessons to be learned from external experience rather than on telling Chinese stakeholders what they already know about their own political economy. What we seek to do is to introduce this international experience and apply it to the Chinese context. Of course, it is true that the Chinese political context is unique. However power system physics and the economic principles on which power market reform are based are not. Indeed many opponents of the power market reforms that have swept through the US, Europe, Australasia and South America initially opposed reform with arguments about the uniqueness of their national power sectors. This is not to say that power sector reform is straightforward or guaranteed to be successful. While reform has been attempted in many countries, it has been ‘successful’ in only a subset of those jurisdictions that have attempted it and it remains a controversial work in progress in every jurisdiction where it has been tried. Indeed at one point, there were good arguments for saying that only in the UK (and perhaps, Norway) has a comprehensive power market reform been ‘successful’, if by comprehensive we mean a reform which created competitive wholesale and retail markets and introduced effective incentive regulation of transmission and distribution networks.

1.2 The Structure of the Book The book is divided up into five chapters. The following sections of this chapter go on to examine the scale and scope of the Chinese electricity supply sector focusing on its capacity, carbon impact, roll-out of renewables and energy demand. We then discuss the structure and organisation of the industry, and introduce the recent history of the sector and the background to the 2015 power sector reforms. Chapter 2 introduces 14 different electricity reform elements from international experience. Under each of these reform elements we will discuss its theoretical significance, general reform experiences with it and its application in the Chinese context. Our motivation is to explain how China might bring down high industrial price of electricity that it had in

1  An Introduction to the Chinese Power System and Its Reform 

3

2014 (prior to the current reform). We identify four promising sources of price reduction: the introduction of economic dispatch of power plants, rationalisation of electricity transmission and distribution, reduction of high rates of investment and rebalancing of electricity charges towards residential customers. We draw out some overall lessons and identify some important points for future research into Chinese power market reform. Chapter 3 aims to go deeper in our discussion of power sector reform by examining China’s most economically significant province: Guangdong. Power market reform in China is being promoted primarily at the provincial level, with the setting up of provincial wholesale power markets. We look at the operation of the pilot wholesale power market in Guangdong in the light of international experience. We discuss how the power market pilot is working in Guangdong and the extent to which the current market design is in line with successful power markets we see elsewhere. We examine the evidence on whether the market reform has successfully brought new players into the electricity system in Guangdong. We consider the effects of the reform on the operational and investment decisions of firms in the sector. We conclude with several lessons for the Chinese government’s ongoing power sector reform programme. Chapter 4 seeks to explain how successful power market reform has worked in practice, drawing on the many questions we have received over the last four years from Chinese stakeholders seeking to understand the British experience with electricity supply sector liberalisation. We begin by discussing the components of the price of industrial electricity in Great Britain, as an example of a fully reformed electricity market, where the market is roughly comparable in size to a typical Chinese province. We go on to discuss the key actors in the liberalised electricity system in Great Britain, before unpacking each of the components of the price. We discuss the market-determined elements first, then introduce and discuss the regulated elements of the price and finish with the central government-determined price components. Our discussion covers the determination of the wholesale price, the retail margin, transmission charges, system balancing charges, distribution charges and environmental levies and taxes. In each of these cases we discuss the process by which they are determined (led by the market, the regulator, the central government or

4 

M. G. Pollitt

more than one) and the specific lessons for China. We conclude by emphasising some of the high-level lessons on electricity price determination for China. Chapter 5 briefly draws out some reflections from the three previous chapters on the future prospects for the reform. We discuss the optimal sequencing of reform and reflect on the latest progress being made with the reform. We conclude the chapter by separately discussing key positive messages for Chinese policy-makers, regulators, generators, retailers and grid companies involved with reform.

1.3 T  he Scale and Scope of the Chinese Electricity Supply Industry in 2015 Before embarking on any discussion of appropriate reform steps to take in the Chinese electric power sector, it is important to acknowledge the achievements of the sector under public ownership and state direction.2 The country has been fully electrified and the technical losses in transmission and distribution were only 5.8% by 2013. The sheer scale of the sector by 2015 is illustrated in Fig. 1.1, which shows the distribution and quantity of capacity type and generated electricity. Figure 1.2 shows the enormous physical build rate and financial investment involved (around $120 bn in 2015). Electrification has kept pace with sustained high rates of demand growth and the sector is self-financing (unlike in India). This has been an impressive engineering undertaking by global standards. In 2014 the Chinese electricity sector had 4 million employees.3 However, China’s rapid growth of electricity demand (at 8.6% p.a. from 2008 to 2014) has appeared to moderate recently (1% in 2015), as can be seen in Fig.  1.3. If this is a sustained ‘new normal’ for power demand, then the rate of investment in new capacity needs to also slow down. In 2015 China was the world’s largest electricity producer (24% of global production), the second largest producer of non-hydro renewable  See Yu (2014).  China Electricity Council (2015).

2 3

1  An Introduction to the Chinese Power System and Its Reform 

Capacity Type in GW (2015) 27.17 , 1.82% 66.03 , 4.43%

4.34 , 0.29%

42.18 , 2.83%

0.09 , 0.01%

Coal

130.75 , 8.77%

Hydro Wind Gas

319.54 , 21.44%

Solar

900.09 , 60.40%

Nuclear Oil Others

Generation Type in TWh (2015) 166.9, 2.99% 185.6, 3.33%

171.4, 3.07% 39.5, 0.71%

4.2, 0.08%

0.1, 0.00%

Coal Hydro Wind Gas

1112.7, 19.95%

Solar 3897.7, 69.88%

Nuclear Oil Others

Fig. 1.1  The size of the Chinese electricity sector. (Source: NBS (2016))

5

6 

M. G. Pollitt

Capacity Under Construction (GW) 90 80 70 60 50 40 30 20 10 0 2008

2009

2010 hydro

2011

2012

coal

nuclear

2013

2014

2015

wind

Annual Investment in the Power System (Bn US Dollars) 80 70 60 50 40 30 20 10 0 hydro

coal

nuclear

wind

solar Power transmission and distribution

Power generation

2010

2011

2012

2013

2014

2015

Fig. 1.2  The scale of power plant capacity in construction. (Source: China Electricity Council (2015))

7

1  An Introduction to the Chinese Power System and Its Reform 

Electricity consumption (TWh)

6000 5500 5000 4500 4000 3500 3000

2008

2009

2010

2011

2012

2013

2014

2015

Fig. 1.3  China’s electricity consumption slowing down? (Source: CEC 2013 website, Available at: http://www.cec.org.cn/guihuayutongji/tongjxinxi/niandushuju/2013-04-19/100589.html)

energy (17% of global production) and the largest producer of coal (3.5 billion tonnes of coal a year or 47% of global production).4 Around 45% of Chinese coal production is consumed in its power sector and 65% of all its electricity comes from coal.5 China’s coal-based electricity sector alone produces from 7 to 9% of global carbon dioxide equivalent emissions,6 and at least one-third of China’s domestic emissions. As  International Energy Agency (2014). China PRC: Electricity and Heat for 2014, Retrieved from https://www.iea.org/statistics/statisticssearch/report/?country=China&product=electricityandheat 5  The data are from China Energy Statistics from NBS (2015). Moreover, China’s coal is depleting fast, although its reserves may seem to be high. The reserves-to-production ratio has reduced to 35 years due to the rapid increase in coal production in the past 30 years, while the comparable number is 250 years in North America, nearly 500 years in Russia and 100 years in India. In addition, the heavy burden of coal transportation (from mines to power plants) also poses challenges for the railway transportation system in China. The share of coal transportation in total rail transportation increased from 41.4% in 2000 to 50.6% in 2011, while the distance it travelled from mines to power plants increased from 548 km in 1990 to 642 km in 2010. See Xu (2017, pp. 30, 32). 6  There are various figures we can calculate for this. Alva and Li (2018) report China’s coal-fired power plant emissions as being 11.1% of total CO2 emissions in 2016. Using total CO2 emissions from IEA (2018) this gives 3587 m tonnes for the power sector. The total figures on power plant emissions will be higher if other power fossil fuel plants and non-CO2 emissions are included. 4

8 

M. G. Pollitt

Figs.  1.4 and 1.5 indicate this rise to global prominence is relatively recent and has occurred extremely rapidly since 2004. These figures provide the backdrop to the significant international interest in the Chinese power market and its reform. As we describe later, economic reform of the power sector has been ongoing since 1985. However, under the current Five Year Plan (FYP13: 2016–20) China’s electricity sector is undergoing a major transition from 25%

20%

15%

10%

5%

19 9 19 0 9 19 1 9 19 2 93 19 9 19 4 9 19 5 9 19 6 9 19 7 9 19 8 9 20 9 0 20 0 0 20 1 0 20 2 0 20 3 0 20 4 0 20 5 0 20 6 0 20 7 0 20 8 0 20 9 1 20 0 1 20 1 1 20 2 1 20 3 1 20 4 15

0% RES-solar

RES-Wind

RES-Others

Fig. 1.4  China’s global share in the fuel consumption for its power generation. (Source: BP’s the Statistical Review of World Energy 2016, Available at: http:// www.bp.com/en/global/corporate/energy-economics/statistical-review-of-worldenergy/downloads.html)

Global CO2e figures are published periodically and electricity and heat figures for emissions are often combined for China. Thus IEA (2018) reports total CO2 emissions for electricity and heat in China as 4358 m tonnes in 2016. We assume total GHG emissions of 49,300 in 2016 (see Olivier et al. 2017). The higher 2016 emissions figure would be 9% of world GHG emissions and is the figure quoted by some; the lower 3587 m tonnes figure is closer to 7% of the world GHG emissions total.

1  An Introduction to the Chinese Power System and Its Reform 

9

60% 50% 40% 30% 20% 10% 0% 1985 1987 1989 1991 1993 1995 1997 1999 2001 2003 2005 2007 2009 2011 2013 2015 Electricity share

CO2 share

Coal production share

Fig. 1.5  China’s global share in CO2, coal and electricity production. (Source: BP’s the Statistical Review of World Energy 2016, Available at: http://www.bp.com/en/ global/corporate/energy-economics/statistical-review-of-world-energy/downloads.html)

a state-managed system to a market price-based one,7 following the publication of the No.9 document of March 2015 which re-launched a new push for ‘power market reform’ in China. This book focuses on the international lessons for China in the light of the current round of power market reforms and paying attention to the particular context of the Chinese electricity system.

 See National Development and Reform Commission (NDRC) PRC China (2016a) An Overview of the 13th Five Year Plan. Retrieved from http://en.ndrc.gov.cn/policyrelease/201612/ t20161207_829924.html and NDRC (2016b) ‘How China’s 13th Five Year Plan Climate and Energy Targets Accelerate its Transition to Clean Energy’. Retrieved from https://www.nrdc.org/ experts/alvin-lin/how-­chinas-­13th-five-year-plan-climate-and-energy-targets-accelerate-its. For commentaries on the current five year plan, please see Ma (2016). China’s 5 Year Plan for Energy, The Diplomat. Retrieved from http://thediplomat.com/2016/08/chinas-5-year-plan-for-energy/ 7

10 

M. G. Pollitt

1.4 T  he Structure and Organisation of the Chinese Power Sector The current structure of the power industry largely dates from 2002 when the previous State Power Corporation (SPC, itself a short-lived successor to the Ministry of Electric Power) was broken up into 11 successor companies (see Xu 2017). There now are a large number of generation companies, which are separate from the main transmission and distribution companies: State Grid Company of China (SGCC) and China Southern Grid (CSG). The largest generation companies (in 2014) were China Huaneng Corporation, China Datang Corporation, China Huadian Corporation, China Guodian and the State Power Investment Corporation. China Guodian was merged with Shenhua Group in 2017 to form China Energy Investment. These are all owned by the central government (via the State-owned Assets Supervision and Administration Commission [SASAC]) and collectively had a market share of installed capacity of about 45% in 2014 (see Chap. 2). The big five companies are among the largest 500 firms in the world, with annual sales ranging from $28 to $82 bn in 2018 (according to the Fortune Global 500 list).8 Their installed capacity is comparable to EdF (which had 136.2 GW in 2014).9 Table 1.1 shows that China had 20 electricity generators with more than 5 GW of capacity in 2014. Note the presence of several large provincial government investors outside the top five (e.g. from Zhejiang, Guangdong and Hubei provinces) and specialist nuclear and hydro generators. Beyond the big five, China Resources Power is part of the China Resources conglomerate and another Global 500 company. All of these companies are majority publicly owned companies. SGCC and CSG are responsible for the transmission and distribution networks within their service territories. They are responsible for billing all customers in their service territories. As such they purchase power on  See https://fortune.com/global500/2019/  See https://www.edf.fr/sites/default/files/contrib/groupe-edf/espaces-dedies/espace-finance-fr/informations-financieres/publications-financieres/rapport-annuels/EDF2014_Essentiels_vdef_va.pdf 8 9

11

1  An Introduction to the Chinese Power System and Its Reform  Table 1.1 Largest generators by capacity in 2014

Company

GW

China Huaneng Group China Guodian Corporation China Huadian Corporation China Datang Corporation State Power Investment Corporation Shenhua Group China China Three Gorges Corporation China Resources Power SDIC Power Holdings Company Zhejiang Provincial Energy Group Company Guangdong Electric Power Development Company China General Nuclear Power Group Beijing Energy Investment Holding Co., Ltd China National Nuclear Corporation Hebei Construction & Investment Group Co., Ltd Jiangsu Guoxin Investment Group Ltd Shenergy Group Hubei Energy Group Co., Ltd Shenzhen Energy Group Co., Ltd Anhui Energy (Group) Company Ltd

151.49 125.18 122.54 120.48 96.67 66.85 50.03 36.52 32.05 27.27 26.95 21.28 17.32 9.19 8.77 7.71 6.76 5.83 5.83 5.55

Source: China Electricity Statistics

behalf of their regulated price customers. There is no separation of retailing and networks in China. SGCC was the world’s fifth largest company with sales of $387 bn in 2018 (in the Fortune Global 500 list) and has global ambitions to build a global interconnected electricity system (see Liu 2015). CSG is also a world top 500 company, with reported revenues of almost $81 bn in 2018 (narrowly behind EdF). In addition there are around 14 other transmission and distribution companies, several of

12 

M. G. Pollitt

whom remain vertically integrated.10 These companies mainly serve remote areas that were only ever, if at all, weakly connected to SGCC and CSG.  The largest of these companies is the Inner Mongolia Power Corporation which serves western Inner Mongolia and is a vertically integrated utility.11 SGCC and CSG are owned by the central government (via SASAC), while the other companies tend to be owned by their provincial or sub-­ provincial levels of government, which has helped preserve their independence. There are also a number of government-owned supply chain companies, specialising in power plant construction and maintenance. These include China Energy Engineering Corporation (CEEC)12 and Sinohydro. However, while the government owns most of the assets of the power sector, there is a significant amount of competition between branches of government (especially between the central government and the provincial governments) and cross-shareholding between the companies, meaning that the effective locus of control is not with the central government. Government investment vehicles such as provincial investment companies are concerned with the profitability of their investments and are often seeking profitable opportunities rather than promoting development per se; however, ultimately they are subject to political control. The largest electricity companies are major employers; in 2014 the five largest generators employed 615,000, CSG 303,000 and SGCC 1.7 million.13 In addition, the electricity industry consumes 60% of the output of the Chinese coal sector.14 China’s coal industry involves a further ten Fortune Global 500 companies. This indicates that any reform of the power sector which substantially affects the use of local coal will also have a serious knock-on effect in the coal sector. The coal sector employed 4.3 m employees in 2014.  See http://www.gcis.com.cn/china-insights-en/china-infographics-en/221-china-independentgrid-companies-and-their-geographical-coverage 11  For more of the history of this, see Xu (2017). 12  CEEC is a Fortune Global 500 company. Thus the Chinese electricity sector has 9 firms in the Fortune Global 500 in 2019: 2 grid companies, 6 of the generators and 1 supply chain company. 13  China Electricity Council (2015). 14  2017 figure. Source IEA (2019, p.IV22). 10

1  An Introduction to the Chinese Power System and Its Reform 

13

1.5 T  he History of Power Sector Reform in China China has embarked on a prolonged process of electricity reform since 1985, as outlined in Fig. 1.6 and Table 1.2. Up until 1984 there was no private involvement in electricity generation; however, after 1985 and in common with other developing countries facing power shortages China

2015 2014

Opinions on the deepening of Power Sector’s Reform T&D rate reform pilot in Shenzhen

2009

Confirmed direct trade reforms

2003 2002

Established The State Electricity Regulatory Commission Separate generation from grid

Stage 3: Comprehensive reforms

Stage 2: Separation of government functions from business activities 1997 1995

Established State Power Corporation Alleviated nation-Wide power shortages

Stage 1: Fund-raising promotion after electricity shortages

1985

Promoted fund raising and investment Empowered provincial power bureaus New tariffs

Fig. 1.6  Reform timeline for electricity sector. (Source: An et al. (2015, p. 6))

14 

M. G. Pollitt

Table 1.2  Reform timeline for electricity sector in China 1980–1984

1985–2001

2002–Present

Industrial structure

Vertical integration

Vertical integration

Ownership

Predominantly central government owned

Dispatch

Economic dispatch based on total embedded cost

Central and provincial government ownership. Increasing private investment in generation Equal shares dispatch

Unbundled generation and transmission and distribution (2002) Central and provincial government ownership, declining share of private investment

Wholesale generation pricing

Internal transfer prices

Investment recovery based on financial lifetime (1985) Investment recovery based on operational lifetime (2001)

Equal shares dispatch; pilot projects for energy-efficient dispatch (2007) Benchmark price (2004) Fuel price-wholesale price co-movement (2004)

Source: Kahri et al. (2013, p. 362)

did allow multiple public investors and private investors into the power generation sector to help alleviate shortages of electricity. The most significant recent reform was the reorganisation of the power sector in 2002 (following Document No.5).15 This saw the separation of generation and transportation/retailing of electricity and represented the most significant structural change to the industry in the modern era. It also saw the creation of two grid companies: State Grid Company of China and China Southern Grid (covering five southern provinces). A regulatory agency for electricity was created and there was an expectation that China would embark on the pathway of the standard international reform model with  This followed the publication of State Council ‘power system reform program’ (the No.5 Document) in December 2002. Source: China5e Research Centre (2016). (Source:中国能源网研 究中心, 中国电改试点进展政策研究与建议). 15

1  An Introduction to the Chinese Power System and Its Reform 

15

the creation of competitive wholesale power markets and regulated network tariffs. However, this process stalled around 2007. Generators continued to receive regulated prices for their power and network tariffs were not separately identified and all customers bought power from their local transmission and distribution monopoly. Thus, transmission, distribution and retail continue to be 100% vertically integrated. There remains significant support for the large grid companies as enabling effective response to large natural disasters (e.g. the Wenchuan earthquake in 2008) and as a counterbalance to the monopoly power of global equipment suppliers such as GE and Siemens.16 It is worth noting that other energy sectors in China, such as oil and gas state-owned companies, were also liberalised during 1990s. Lin (2008) points out that the Chinese central reformers of oil and gas industries recognised the need to depart from the decentralised approach to industrial governance as early as 1993.17 However, the reform process had to wait until key domestic interest groups were weakened by macroeconomic disequilibria and global price shocks in the latter half of 1990s. Thus, the Chinese electricity sector reforms, similar to other global reform experiences, take a long period of time and rely on national cross-sectoral learning. The previous reforms of the energy prices, resource taxes and subsidies have also paved way for this round of electricity market reform, which has made great progress in assigning a more significant role for the market in allocating resources in China since 1984 (see Mou 2014; Lin and Ouyang 2014; Paltsev and Zhang 2015; Zhang 2014). Moreover, a new wave of comprehensive reforms was launched by the Chinese leadership in November 2013, and the electricity sector is under the government spotlight due to its important role in helping China’s transition to a low-­carbon economy and in addressing local air pollution. As Table 1.2 shows, the regulated prices received by individual power plants have become more sophisticated over time in adjusting to wholesale fuel prices (particularly the price of coal).18 However, dispatch is not done on a least cost (merit order basis), but on an equal shares basis. This involves plants of a similar vintage being allocated an equal number of annual  See Xu (2017, pp. 157, 258 and 303).  See Lin (2008, 2014). 18  For more details on the substance of the 2002 transition, see Wang and Chen (2012). 16 17

16 

M. G. Pollitt

running hours and being dispatched on a daily basis in line with the need to achieve an equal number of total running hours. We return to this later. The latest round of power market reforms began in March 2015, promoted by the publication of the CPC Central Committee and State Council No.9 Document of March 2015 (summarised in Table  1.3). This document foresees a renewed push to establish competitive wholesale and retail electricity markets especially for industrial electricity customers. It is supported by a number of ongoing market pilot projects. For the purposes of our subsequent discussions it is worth noting that the seven policy goals in the No.9 document relate to elements of power market reforms undertaken in other countries. Policy Goal No.1 relates to unbundling of regulated network charges from competitive segment charges. Policy Goals No.2–No.5 are around the creation of a competitive wholesale market with sufficient vertical separation of generation and retail from the natural monopoly elements. Policy Goal No.6 targets the efficient allocation of scarce transmission capacity. Interestingly, the wording of the No.9 document emphasises the ‘trading’ of electricity. International reform experience emphasises the use of market mechanisms and the harnessing of competition to allocate scarce resources. It also emphasises ‘separation’, the idea that the roles of various actors within the sector need to be clearly defined and in particular that boundaries need to be strictly drawn between competitive and monopoly activities, with a key role for regulation in ensuring non-discriminatory access and implementing incentive regulation. Additionally, under the Policy Goals No.2, 4 and 6, the No.9 document emphasises the significance of ancillary services and stresses the establishment and improvement of purchasing mechanisms for ancillary services.

1.6 W  hat Motivated the 2015 Power Sector Reform? There is a clear motivation for the timing of the current round of reforms, which has come across in our discussions with Chinese stakeholders. The US had been enjoying a period of falling power prices driven by the rising amounts of cheap natural gas available within the US, due to the shale gas boom. This had reduced the industrial price of electricity in the

Sources: ‘Deepening Reform of the Power Sector, Document No. 9, March 21, 2015’, China State Council (2015) and China5e Research Institute (2016, pp. 4–5)

Policy Goal No.7 Reinforcing electricity safety, scientific supervision and an integrated power planning system

Policy Goal No.4 Establishing independent electricity trading institutions and a fair and regulated trading platform Policy Goal No.5 Steadily reforming power sales side and distribution Policy Goal No.6 Enhancing fair access to power grid and power transmission

Policy Goal No.3 Reforming power generation, power utilisation and the current market mechanisms

State Council China and CPC Central Committee Implementing Opinions on Document No.9 issued the ‘Opinions on Further Deepening Implementation Opinions on Promoting Transmission-­ Power Sector Reform’ (Document No.9) in March Distribution Price Reform 2015. There are two main stages for this round Notification of perfecting formation mechanism of of electricity reform in China. In the first stage trans-provincial and trans-regional power trading (from March to June 2015), NDRC and other prices related governmental agencies announced five Implementation Opinions on Promoting Power Market supporting documents. In the second stage Construction (November 2015), NDRC and NEA further issued Notification of Perfecting Power Emergency Response another six supporting documents. These Mechanism supporting documents provide the practical and Comprehensive City Pilots of Managing Powerguidance for implementing the seven main policy Demand Side goals set in the Document No.9, which cover the Implementation Opinions on Orderly Releasing Plans issues of electricity price, power trading system, of Power Generation and Power Utilization wholesale side design, power grid and Implementation Opinions on Establishing Power governmental supervision Trading Institutions and Their Normative Operation Electricity ancillary services in China have long been provided by grid-connected power plants. Document No.9 changes this situation by Implementation Opinions on Promoting Power-Sales establishing a new ‘shared responsibility’ Side Reform mechanism. This ‘shared responsibility, shared gains’ mechanism improves the original Guidance Opinions on Improving Power Operation compensation mechanism, and welcomes user Adjustment to Facilitate Multiple and Full participation in ancillary services by contracting Development of Clean Energy with either generator companies or the grid. In Guidance Opinions on Reinforcing and Regulating March 2015, the supplement policy document— Supervision and Management of Coal-Fired Guiding Opinions on Improving Electric Operation Self-Generation Power Plants and Regulation to Promote Greater and Fuller Use Supervision and Examination Procedures for Pricing of Clean Energy—was published, which aims to Costs of Power Transmission and Distribution (Trial) advance the ancillary services and promote renewable energy consumption at the same time

Policy Goals No.1 Promoting electricity power pricing mechanisms Policy Goal No.2 Reforming power trading systems and refining market-­oriented trading systems

Reform process (mentioned in Document No.9)

Supporting documents

Key policy goals

Table 1.3  Document No.9 of March 2015 and stated reform process

18 

M. G. Pollitt

Table 1.4  Electricity price and fuel input price differential with US in 2014 Industrial electricity Price (US $/kWh) In 2014 US China China minus US

0.0710 0.1068 0.0358 (50% higher)

Coal price for generation (US $/kWh) in 2014

Gas price for generation (US $/kWh) in 2014

Residential electricity price (US $/kWh) in 2014

0.0241 0.0384 0.0143

0.0159 0.0778 0.0619

0.1252 0.0908 −0.0344 (27% lower)

Notes: Chinese prices include VAT tax, see http://cn.manganese.org/images/ uploads/board-documents/8._2015_AC_-_Xizhou_Zhou-CN.pdf (p. 20) Source: Chinese data from Chinese government website (http://zfxxgk.nea.gov.cn/ auto92/201509/t20150902_1959.htm) and US data from EIA website Exchange rate: 6.1428 RMB/USD

US. Meanwhile in China power prices had remained relatively high due to the fact that all prices, both for generators and for final customers, were fixed. This situation meant that the industrial price of electricity in China was significantly higher than that in the US by 2014, as illustrated in Table 1.4. Table 1.4 shows two major differences with the US—residential prices are lower than that in the US and lower than industrial prices and fuel input prices are significantly higher in China. The higher industrial price is not fully explained by higher marginal fuel prices (where gas is the fuel of choice in the US and coal is the fuel of choice in China). Higher marginal fuel prices in 2014 (coal in China minus gas in the US) only explains 63% of the differential, meaning that 37% (or 12% of the 2014 industrial price) is not explained by fuel cost differentials, though some of the price is explained by the higher general value added taxation on the sector. However, given the lower unit labour and unit capital costs in China, we might expect Chinese non-fuel costs to be lower than that in the US. The reform of residential tariffs in China is difficult due to the political economy of raising power prices to cost-reflective levels. Moreover, the cross-subsidy from higher industrial electricity prices to lower residential prices can be viewed as a way of improving the efficiency of energy-intensive companies (see Sun and Lin 2013; He and Reiner 2016; Zhang 2014).

1  An Introduction to the Chinese Power System and Its Reform 

19

In Chap. 2, we discuss how power sector reform based on the lessons of international experience could eliminate the unexplained price differential between China and the US and reduce the price of industrial electricity in China by 12% or more.

References English Alva, H. A. C., & Li, X. (2018). Power sector reform in China: An international perspective. Paris: IEA Publications. An, B., Lin, W., Zhou, A., & Zhou, W. (2015). China’s market-oriented reforms in the energy and environmental sectors. Paper presented at the Pacific energy summit. Retrieved from http://nbr.org/downloads/pdfs/ETA/PES_2015_workingpaper_ AnBo_et_al.pdf BP. (2016). Statistical review of world energy 2016, BP. CEC (China Electricity Council). (2013). Electricity industry statistics summary 2012. Internal report, China Electricity Council. He, X., & Reiner, D. (2016). Electricity demand and basic needs: Empirical evidence from China’s households. Energy Policy, 90, 212–221. International Energy Agency [IEA]. (2014). China PRC: Electricity and heat for 2014. Retrieved from https://www.iea.org/statistics/statisticssearch/report/?c ountry=China&product=electricityandheat International Energy Agency [IEA]. (2018). CO2 emissions from fuel combustion: 2018 edition. Paris: OECD. International Energy Agency [IEA]. (2019). Coal information: 2019 edition. Paris: OECD. Kahri, F., Williams, J. H., & Hu, J. (2013). The political economy of electricity dispatch reform in China. Energy Policy, 53, 361–369. Lin, K.  C. (2008). Macroeconomic disequilibria and enterprise reform: Restructuring the Chinese oil and petrochemical industries in the 1990s. The China Journal, 60, 49–79.

20 

M. G. Pollitt

Lin, K. C. (2014). Protecting the petroleum industry: Renewed government aid to fossil fuel producers. Business and Politics, 16(4), 549–578. Lin, B., & Ouyang, X. (2014). A revisit of fossil-fuel subsidies in China: Challenges and opportunities for energy price reform. Energy Conversion and Management, 82, 124–134. Liu, Z. (2015). Global energy interconnection. San Diego: Elsevier. Ma, T. (2016). China’s 5 year plan for energy. The diplomat. Retrieved from http://thediplomat.com/2016/08/chinas-5-year-plan-for-energy/ Mou, D. (2014). Understanding China’s electricity market reform from the perspective of the coal-fired power disparity. Energy Policy, 74, 224–234. National Development and Reform Commission PRC China. (2016a). An overview of 13th five year plan. Retrieved from http://en.ndrc.gov.cn/policyrelease/201612/t20161207_829924.html National Development and Reform Commission PRC China. (2016b). How China’s 13th five year plan climate and energy targets accelerate its transition to clean energy. Retrieved from https://www.nrdc.org/experts/alvin-lin/ how-chinas-13th-five-year-plan-climate-and-energy-targets-accelerate-its Olivier, J. G. J., Schure, K. M., & Peters, J. A. H. W. (2017). Trends in global CO2 and total greenhouse gas emissions. Summary of the 2017 report, The Hague: PBL Netherlands Environmental Assessment Agency. Paltsev, S., & Zhang, D. (2015). Natural gas pricing reform in China: Getting closer to a market system? Energy Policy, 86, 43–56. Sun, C., & Lin, B. (2013). Reforming residential electricity tariff in China: Block tariffs pricing approach. Energy Policy, 60, 741–752. Wang, Q., & Chen, X. (2012). China’s electricity market-oriented reform: From an absolute to a relative monopoly. Energy Policy, 51, 143–148. Xu, Y. C. (2017). Sinews of power: Politics of the state grid corporation of China. Corby: Oxford University Press. Yu, H. (2014). The ascendency of state-owned enterprises in China: Development, controversy and problems. Journal of Contemporary China, 23(85), 161–182. Zhang, Z. X. (2014). Energy prices, subsidies and resource tax reform in China. Asia and the Pacific Policy Studies, 1(3), 439–454.

Chinese China Electricity Council. (2015). Annual electricity statistics 2015. Available at: http://www.cec.org.cn/

1  An Introduction to the Chinese Power System and Its Reform 

21

China State Council. (2015). 关于进一步深化电力体制改革的若干意见. Deepening Reform of the Power Sector, Document No. 9, March 21. Retrieved from http://www.ne21.com/news/show-64821.html China5e Research Centre. (2016). 中国电改试点进展政策研究与建议. The Policy Study of China’s Power Sector Reform Pilot Project Development and Recommendations. China5e Research Centre Report, Beijing. National Bureau of Statistics (NBS). (2015). China energy statistical yearbook 2015. Beijing: China Statistics Press. National Bureau of Statistics (NBS). (2016). China electric power yearbook 2016. Beijing: China Electric Power Press.

2 Lessons for China from International Experience of Power Sector Reform In collaboration with Chung-Han Yang and Hao Chen

2.1 Introduction Looking across the world at electricity reform we can identify 14 reform elements that form part of a modern power market reform. We take 11 of these from Paul Joskow (2008)1 who identifies 11 key components of successful processes and supplement these with 3 additional reform elements appropriate to a low carbon transition (based on Pollitt and Anaya 20162), which involves currently subsidised (but lower carbon) generation technologies. We group the main reform components into four general areas: market restructuring and ownership changes, supportive secondary market arrangements, appropriate economic regulation and efficient promotion of low emission technologies. To these four general areas Joskow adds appropriate transition mechanisms to recognise the fact that power market reform is complex and that a  Points 1–10 and 14 below are from Joskow (2008).  Points 11–13 below are from Pollitt and Anaya (2016).

1 2

© The Author(s) 2020 M. G. Pollitt, Reforming the Chinese Electricity Supply Sector, https://doi.org/10.1007/978-3-030-39462-2_2

23

24 

M. G. Pollitt et al.

successful transformation cannot be achieved in one step. We use these to organise our discussion of the Chinese case described later. Successful power market reform in a low emissions context, according to Joskow (2008) and Pollitt and Anaya (2016), must therefore take due account of the following: Market restructuring and ownership changes: (1) vertical separation of competitive elements (generation and retail) from natural monopoly networks, (2) sufficient horizontal restructuring of generation to create a competitive wholesale market, (3) the creation of wide area independent system operators and (4) privatisation of monopolies. Supportive secondary market arrangements: (5) creation of spot and ancillary services markets to support real-time balancing of the system, (6) participation of demand side in wholesale electricity markets and (7) regulated third-party access to, and efficient allocation of, scarce transmission capacity. Appropriate economic regulation: (8) unbundling of regulated network charges and competitive segment charges, (9) mechanisms to ensure competitive procurement of wholesale power for regulated final customer groups and (10) the creation of independent regulatory agencies to regulate monopoly network charges and monitor competitive segments. Efficient promotion of low emission technologies: (11) competitive procurement processes for low carbon generation, with some exposure to wholesale price variability, (12) cost reflective access terms for renewables and (13) appropriate pricing of environmental externalities (both carbon dioxide and other atmospheric pollutants, such as sulphur dioxide). And finally, all good power market reforms (and, indeed, significant economic reforms more generally) involve: (14) appropriate transition mechanisms.

2  Lessons for China from International Experience of Power… 

25

The aim of this chapter is to lay out the extensive reform experience of European countries (which began in 1990 in the UK), as well as other reforming countries (in particular, the US) and apply it to the situation of China. The European Union (EU) legislative process consisted of three electricity directives (1996, 2003 and 2009), which have successively opened up the EU electricity market, in line with Joskow’s model for successful reform (see Jamasb and Pollitt 2005; Pollitt 2009). These have introduced a competitive wholesale market, regulated third-party access to transmission and distribution networks, legal separation of retail businesses, choice of retailer for all customers, unbundling of transmission and distribution businesses from the rest of the sector, regulated cross-­border trading and independent regulatory authorities. The process has been slow and different European countries have at times moved at very different speeds in implementing reform, but the overall progress has been remarkable. Since the start of the reform process in the EU there has also been a significant push towards a low carbon electricity system with a large emphasis on increasing the share of renewables in the electricity system and the introduction of a cap on emissions from the electricity sector as well as the introduction of the EU Emissions Trading System3 (EU ETS). China is now undergoing its own renewable and low carbon electricity transition, with a remarkable growth of renewable and nuclear energy (from 17% to 30% over the period 2003–18)4 and significant moves towards a national carbon market.5 This poses new challenges for the reform process around how to successfully integrate renewables into wholesale energy and ancillary services markets, how to facilitate appropriate levels of network access for renewables, appropriate mechanisms for financing renewables and whether there are implications for the remuneration mechanisms for fossil fuel power plants in the presence of large amounts of renewables. Reform points 11–13 mentioned earlier are much less well developed globally (and less supported by empirical evidence). Advanced jurisdictions (such as the UK, Germany, California  Newbery (2016) and Ellerman et al. (2016).  BP (2019), share of hydro, other renewables and nuclear in total electricity production. 5  See China State Council (2015). 3 4

26 

M. G. Pollitt et al.

and New York) are currently experimenting with different mechanisms to support their low carbon transition (Pollitt and Anaya 2016).6 This chapter aims to collect information relevant to all of the 14 points identified earlier, and drawing on our discussions with Chinese stakeholders, comment on what China is doing under each of these reform elements. It will seek to assess progress with reform and also what China is learning about how the reform model needs to be adapted for its own particular circumstances. While the outline of a successful reform model may be easily stated, the details vary from jurisdiction to jurisdiction. Thus electricity reform in Germany is very different from the UK and that in California is different from that in New York. In what follows, we aim to draw out what the particular lessons from the reform model outlined earlier are for China. Most importantly, we seek to identify what are the key institutional problems to be overcome in bringing about a successful electricity reform transition in the world’s most significant electricity system. The paper proceeds as follows. We discuss each of the 14 reform elements under the groups of headings that we identified earlier. Under each of these reform elements we will discuss its theoretical significance, general reform experiences with it and its application in the Chinese context. We conclude with some overall lessons and identify some next steps, which we will go on to discuss more fully in subsequent chapters.

2.2 Market Restructuring and Ownership Changes 2.2.1 V  ertical Separation (1) and Horizontal Restructuring (2) 1. vertical separation of competitive elements (generation and retail) from natural monopoly networks; 2. sufficient horizontal restructuring of generation to create a competitive wholesale market;  See Pollitt and Anaya (2016).

6

2  Lessons for China from International Experience of Power… 

27

2.2.1.1  Theoretical Significance The electricity sector consists of different vertically related segments with different cost and innovation characteristics. These give rise to different minimum efficient sizes of firms in relation to the relevant market segment. These segments are electrical equipment, generation, transmission, distribution and retailing. Electrical equipment is a competitive input sector, subject to global competition. Generation can be organised into wide area markets where generating firms, which could consist of a single power plant, can compete to provide electricity at lowest cost and to invest to meet future power demand. Transmission has natural monopoly characteristics in the operation of given collections of assets in particular areas. Distribution is a local natural monopoly (often very local) in the operation and investment in lower voltage networks. Retailing is a potentially competitive activity, which involves contracting for power and metering and billing final electricity customers. Individual retail firms can operate over wide areas or concentrate on particular geographies. Each of these activities can have very different minimum efficient scales, risk profiles and very different dominant logics among their management teams. Generation and retailing require significant marketing and trading activity, while transmission and distribution are engineering-led activities. Generation and retailing are higher risk investments, while transmission and distribution are much lower risk investments. Differences in the characteristics of the different vertical segments of the electricity sector argue strongly for vertical separation. Where network monopoly segments remain integrated with competitive generation/retail activities it is necessary for access to these segments to be priced on a non-discriminatory basis, so that any competing generation or retail firms with identical network access requirements are charged the same access charge in order not to distort competition between them. The generation market is potentially competitive in all but the smallest electricity systems.7 However such competition depends on the existence of sufficient firms in the price-setting part of the market. Thus a large  See Bessant-Jones’ (2006) discussion of the success of wholesale electricity market in Guatemala with a total capacity of 1.875 GW in 2002. 7

28 

M. G. Pollitt et al.

number of base load power plants will not discipline the price in the market at peak times where there are only a small number of firms with peak power capacity. If there are only a small number of firms with price-­setting plants, then collusion is likely. It is also the case that if one firm has the ability to strategically withdraw capacity such as to leave the market without sufficient capacity (as measured by the Residual Supply Index), then that firm can exercise market power to generally raise prices.8

2.2.1.2  General Reform Experience Reform experience across the world has involved significant vertical separation. The EU electricity directives (1996, 2003 and 2009) have specified the creation of competitive generation and retail markets with full legal unbundling from the monopoly transmission and distribution networks.9 This has over time led to divestment of transmission and distribution businesses and the creation of separate generation-retail companies. There has been a progressive opening up of the retail market to competition, starting with large industrial users of electricity, then all non-­domestic customers and finally domestic customers. In the US there has been a similar process of reform, with a notable absence of privatisation, because most of the industry was already in the private sector.10 This has involved many individual states forcing significant generation asset sales by incumbent integrated utilities, the expansion of regional generation markets and the gradual extension of retail competition from large industrial users to smaller users. Where vertical disintegration has not been pursued aggressively there are some well-documented cases of continuing abuse of market power by incumbent monopoly network companies against their competitors in the competitive segments of the industry, such as in Chile11 and Germany.12

 See Rahimi and Sheffrin (2003) and Sheffrin (2002).  See Jamasb and Pollitt (2007) and Pollitt (2008b). 10  See Joskow (2003). 11  See Pollitt (2005). 12  See Bergman et al. (1998, pp. 158–162). 8 9

2  Lessons for China from International Experience of Power… 

29

The introduction of competitive wholesale power markets that move away from regulation of the generation segment of the electricity sector does pose a significant risk in any power market reform. The early wholesale market in Great Britain following privatisation was characterised by two large firms setting the price 90% of the time.13 This gave rise to tacit collusion between the two firms, which resulted in the need for first price regulation in the wholesale market and then forced sales of generation assets to create a more competitive market. Significant problems with the Residual Supply Index existed in the Californian electricity market in the run-up to its power crisis of 2000–01.14 However, experience has suggested that market power risks can be mitigated by allowing generators to sign longer-term contracts with suppliers. Long-term contracts are much more potentially competitive than spot-market contracts, because they can be signed with new entrants prior to entry. It was the failure to allow long-term contracting that significantly contributed to the Californian electricity crisis.15 Short-term market power can also be significantly mitigated by ‘market abuse’ regulation, which limits the ability of generators to strategically withdraw capacity from the market at short notice to drive up prices.16 The significance of this sort of collusion between fossil fuel generators has declined in many markets with the rise of subsidised must-run renewable generation and the general slowdown in electricity demand growth.

2.2.1.3  Chinese Context China has taken significant and impressive steps to vertically restructure its electricity industry. The most impressive of these was the 2002 reorganisation, which created the seven large companies from one single state-owned company.17 This did effectively separate generation from transmission and distribution. It also created two comparative  See Newbery (2005).  See Sweeney (2002). 15  See Sweeney (2002). 16  For instance in Great Britain, generators license conditions include such a clause. 17  See Cunningham (2015). 13 14

30 

M. G. Pollitt et al.

GWh Distributed per Employee 3 2.5 2 1.5 1 0.5 0

China Southern Grid

State Grid

Fig. 2.1  Relative performance of State Grid and China Southern Grid. (Sources: SGCC (2015) and CSPG (2015))

transmission companies. Figure 2.1 shows some evidence that the smaller China Southern Grid is a more efficient company than State Grid, suggesting the value that China Southern Grid provides as spur to greater efficiency in the sector as a whole. It also created the potential for genuine national competition in generation, drawing on the experience of competition in markets generally, which suggests that five roughly equally sized firms are the minimum requirement to ensure effective competition in a market.18 However the size of all these companies remains a barrier to the emergence of a competitive market in generation, retail and procurement and retail. In terms of assets in 2014, State Grid was four times larger than China Southern Grid and three times larger than the largest generator. State Grid has 1.7 million employees, making it one of the largest corporations in the world by employees and it contains 40% of all the power sector’s employees (including equipment manufacturers).19

 See also Kahrl et al. (2016) and Kahrl et al. (2013).  Source: China Electricity Council.

18 19

2  Lessons for China from International Experience of Power… 

31

Further steps have been taken recently to separate out the network businesses within the two main grid companies in order to clearly identify the regulatory asset base and associated network cost from the power procurement cost, as part of the market pilot projects and in line with Policy Goal No.1 of the No.9 document. This is necessary to allow identification of a non-discriminatory third-party access tariff for network access. Unsurprisingly, State Grid and China Southern Grid exercise significant influence on the supply chain and on the course of power market reform in the electricity sector in China. This is because they continue to integrate transmission, distribution and retailing within their extremely large service areas. They are also absolutely very large companies with significant political influence within the sector, able to influence the speed of reform in the competitive segments. It was clear that after the 2002 reforms China was following the international reform model and that there was a need for further reforms around 2007 to complete the initial separation process. However, following a severe winter in 2008, which caused some power shortages, the grid companies were able to argue that progressing with vertical separation put security of supply at risk and further reforms were halted at that time. Other countries’ experience suggests that the organisation of electricity sector does not require transmission and distribution to be integrated.20 There is also no evidence that security of supply is put at risk by vertical unbundling of transmission, distribution and retailing. Quite the reverse, the evidence is that if anything countries with complete ownership of unbundling of distribution (from both transmission and distribution) have seen improvements in quality of service.21 Transmission systems are subject to regional/national monopoly, whereas distribution can be a local monopoly at the level of the province or municipality. This has the advantage of comparative competition between distribution companies and competition between management teams, increased competition in input markets and increased responsiveness to customer demands (from generators and from electricity customers) for quality of service.  Global trends in transmission system operation arrangements are discussed in Chawla and Pollitt (2013). 21  See the discussion of the experience of New Zealand with ownership unbundling of distribution in Nillesen and Pollitt (2011). 20

32 

M. G. Pollitt et al.

State Grid and China Southern Grid currently have a monopoly of retailing for almost all customers, apart from a number of large electricity users who can self-generate. This will slow the process of introducing competition into the wholesale market where individual generators compete with grid companies with huge power procurement portfolios for final customers. Such competition needs to be non-discriminatory and this depends on the access charge that the grid companies charge correctly reflecting the average cost of transmission and distribution (not including retailing costs). The ability for incumbent network-retailers to reallocate costs between their network and retail businesses can significantly slow the process of competition. They can do this by allocating much of the fixed costs of their retail business to the network business. In the UK the regulator enforced strict asset allocation rules between distribution and retail, within the incumbent regional electricity distribution companies, as the market was opened up to competition. This was because the companies initially tried to allocate 90% of their shared assets to distribution, in order to increase network access charges to new entrants and reduce costs within their own retail businesses. The regulator ruled that only around 75% of these costs could be allocated to distribution (see Domah and Pollitt 2001). As we have already noted there already is substantial dispersion of ownership of generation assets between companies in China. The 2002 reform did lead to a substantial drop in the price paid to generators for coal-fired generation (apparently due to significant competition in construction between generators).22 There is also a substantial emerging surplus of fossil power generation, which suggests that wholesale power prices will not rise (ceteris paribus) as the market for industrial power is opened up to competition. However within particular provinces residual market power is a potential issue, especially where there are transmission constraints which give rise to must-run fossil fuel plants on the system (for grid stability reasons).23  Construction costs for coal-fired generators reportedly fell from 8000 Yuan/kW to 4000 Yuan/ kW, indicating that even if the price of power was regulated the separate generators had the incentive to cut their supply chain costs. 23  For a discussion of the Yunnan pilot in this context, see Feng (2016a). 22

2  Lessons for China from International Experience of Power… 

33

Table 2.1 Benchmark generation tariffs of coal-fired power plants (with FGD) in 2014

Province Beijing Tianjing Hebei North Hebei South Shanxi Inner Mongolia West Inner Mongolia East Liaoning Jilin Heilongjiang Shanghai Jiangsu Zhejiang Anhui Fujian Jiangxi

Province

Electricity tariff (in RMB/ kWh)

Electricity tariff (in US $/ kWh)

Hubei Hunan Guangdong Guangxi Hainan Chongqing

0.4702 0.5269 0.5122 0.4672 0.4888 0.4401

0.0765 0.0858 0.0834 0.0761 0.0796 0.0716

0.0605

Sichuan

0.4607

0.0750

0.0671 0.0666 0.0578 0.0755 0.0720 0.0763 0.0705 0.0715 0.0793

Guizhou Yunnan Shaanxi Gansu Qinghai Ningxia Xinjiang Henan Shandong

0.3791 0.3633 0.4002 0.3329 0.3570 0.2862 0.2620 0.4382 0.4472

0.0617 0.0591 0.0651 0.0542 0.0581 0.0466 0.0427 0.0713 0.0728

Electricity tariff (in RMB/ kWh)

Electricity tariff (in US $/ kWh)

0.3987 0.4085 0.4228 0.4316 0.3887 0.3094

0.0649 0.0665 0.0688 0.0703 0.0633 0.0504

0.3714

0.412 0.4094 0.355 0.4638 0.442 0.469 0.4331 0.4393 0.4872

Note: The exchange rate between US $ and RMB is 6.1428 in 2014 Source: NEA website (http://www.nea.gov.cn/)

Generators receive regulated prices for their generation (except where their generation is being sold in the markets created since 2015). Table 2.1 shows the tariffs for coal-fired generation in 2014. These are often above the final retail price of industrial power in the US ($0.071/kWh), showing that the regulated prices look generous by international standards. The scope for competition in generation exists in China. The issue is the extent to which transmission constraints will allow existing fossil fuel generators to compete with each other for final customers. One issue that is relevant in the wholesale market pilots is that these markets only cover part of both supply and demand. In order for a meaningful market price to emerge in this segment supply and demand curves must be allowed to cross and give rise to single equilibrium price in each trading period. This single market price should be paid by all demanding loads to all

34 

M. G. Pollitt et al.

supplying generators in the market at that price. This means that sufficient amounts of capacity and load need to be in the market such that the amount of generation being allocated by a market mechanism can give rise to meaningful price. The introduction of wholesale markets, in which fossil fuel generators were expected to participate without subsidy, would subject new fossil fuel generation investment to market incentives and reduce over investment in new fossil fuel generation capacity. If the amount of supply and demand in the market is restricted to be less than a market clearing quantity, then the generators in the market can exercise market power and charge high prices. This seemed to be the case in the Yunnan pilot in 201624 and in the Guangdong pilot in 2017 (discussed in the next chapter), where the price determination mechanism was to restrict demand and supply in the monthly contract market and then inversely match and average the highest bids and lowest offers. This gave rise to different ‘market’ prices for each block of power in the market, but also scope for gaming by buyers and sellers. Thus, for example, the lowest cost generator could raise its bid and receive a higher payment.

2.2.2 T  he Creation of Wide Area Independent System Operators (3) 2.2.2.1  Theoretical Significance A system operator is the ‘air traffic controller’ of the electricity system.25 A key job of the system operator is to balance the market in real time on a least cost basis. The larger the control area, the more the system can optimise the use of low cost sources of generation and economise on the holding of reserve capacity (both in the short run and in the long run). Competitive wholesale markets for electricity are usually co-incident with the area of operation of a single system operator. In many liberalisation processes (e.g. in England and Wales) a wide area system operator already existed and it was a straightforward process to move from cost-­ based merit order dispatch to bid-based dispatch. Wide area  See Cheng et al. (2018) and Liu et al. (2019) for discussions of the reform in Yunnan.  See O’Donnell (2003).

24 25

2  Lessons for China from International Experience of Power… 

35

dispatch on the basis of least cost is the key to reducing total system operating costs. Extensions of system operator control areas by merging pre-existing control areas have the effect of increasing wholesale market size and single price areas (this has happened in Great Britain, with the extension of National Grid’s control area to include Scotland as well as England and Wales, and in the US with the extension of PJM’s control area; PJM is the largest organised market in the US). The operational independence of the system operator from generators and from retailers (and from local, provincial and national governments) is important because of the strong link between being physically dispatched and the revenue of individual generators. It is essential that dispatch is in the best interests of the system as a whole rather than the narrow interests of one ownership party (or group of parties) in the system.

2.2.2.2  General Reform Experience The evidence is that system integration and joint system operation and dispatch have significantly reduced costs and improved efficiency. The extension and evolution of independent system operators in the US, bringing together the previous multiple control areas of individual vertically integrated utilities, has reduced costs. PJM’s control area extension has produced significant measured benefits in terms of reducing pricing inefficiencies between previously separated areas.26 System operator dispatch can be successfully conducted on three different bases in liberalised markets. Each of them involves dispatch consistent with least cost power plants being dispatched first. Cost-based dispatch has been practised in Latin America27 and in Ireland.28 This involves plants being dispatched in order of audited marginal operating costs calculated on the basis of known plant operating parameters. This is a good way of restraining market power in smaller markets with large pivotal owners of generation plant. Central price-based dispatch involves  See Mansur and White (2012).  See Newbery (2016). 28  See Pollitt (2008b). 26 27

36 

M. G. Pollitt et al.

dispatching plants on the basis of price-based bids. This is similar to marginal cost-based dispatch except the parameters used are bids submitted by the generators in the day-ahead market. This is the standard dispatch system in US ISOs. Self-dispatch is the practice in the EU.29 This involves generators declaring that they want to be dispatched to the system operator (on the basis of their contractual position), who then has to dispatch them subject to system operating constraints. In theory self-dispatch is more efficient than central dispatch because it can reflect more up-to-­ date information on the operating and demand conditions facing individual plants. In practice it gives rise to the possibility that plants will be dispatched out of merit order. Evidence suggests it is slightly more inefficient than central dispatch.30

2.2.2.3  Chinese Context In China there are six regions of operation of generation system (see Fig.  2.2) and the 2002 reform envisaged moving towards six regional power markets.31 This process has not been completed and dispatch is largely organised at the provincial level with some higher-level regional management of bulk power flows—which are often seasonal—between provinces. Provincial-level dispatch is inefficient and does not fully exploit the large opportunities for trading power across regions (e.g. between Yunnan where this often surplus hydro power and Guangdong where there are much higher marginal production costs).32 Guangdong has three dispatch centres, while Beijing, Tianjin and Hebei operate a joint dispatch centre. The dispatch centres are owned and operated by the relevant grid company, but the annual allocation of hours (for the period 1 January to 31 December) is determined by the provincial government, though the process of its determination is not always transparent or  See Pollitt (2012).  See Sioshansi et al. (2006). 31  See Zhang et al. (2015a). 32  Electricity trading among different provinces will affect the electricity generation from local power companies, thus affecting the provincial GDP. Therefore, this political economy between central government and local government is a barrier for the wider area for electricity trading. See Xu (2017, pp. 141, 176–179). 29 30

2  Lessons for China from International Experience of Power… 

37

Fig. 2.2  Regional and provincial grid control areas in China. (Source: Derived from Wang and Chen (2012, p. 144))

timely (i.e. sometimes after the start of the current year). In addition, there is a national dispatch centre and city- and county-level dispatch centres.33 Dispatch is not currently merit order based and is in need of reform. The provincial dispatch centre is part of the State Grid of China or China Southern Grid (whose provincial areas are also indicated in Fig.  2.2). Plants are dispatched to meet demand on the basis of target annual running hours (+/− 1.5%), subject to priority dispatch for nuclear power plants and renewables.34 There are also monthly dispatch plans. As 33 34

 See Kahrl and Wang (2014).  See Pingkuo and Zhongfu (2016).

38 

M. G. Pollitt et al.

discussed earlier, this implies that on any given day the dispatch schedule will be drawn up on the basis of the cumulative running hours total for the year. Fossil fuel power, which is running further behind its target annual running hours, is more likely to be dispatched first. According to China’s Renewable Energy Law (2010)35 all available renewable power should be dispatched first. In practice renewables are often constrained off by a combination of transmission constraints, a desire to help fossil fuel power plants meet their annual hours target and the fact the renewables are financially expensive (per kWh of dispatched generation). Market pilots have focused on monthly contracts for power and the incentive for generators to participate in such pilots is that if they sell more power in the contract market then this can be used to justify the need for them to be dispatched more than would otherwise be the case by their provincial dispatch centre. Given that allocation of annual hours often occurs after the start of year, holding contracts in the market can justify a higher allocation of hours. The organisation of dispatch is ripe for reform in China. In 2015 around 1.6% of power demand could have been met by renewable generation that was constrained off the system (we discuss curtailment of renewables later in the chapter). This is essentially free electricity (once the investment has been sunk). Academic studies suggest that efficient dispatch might reduce coal demand in China by up to 6% via a combination of reducing lost renewable output and dispatching more efficient coal-fired power plants first.36 What is striking is that dispatch savings alone are actually quite small (though they represent essentially free money left on the table within the existing power system). The environmental savings are large at 0.5% of global CO2e emissions.37 Coal cost  China’s Renewable Energy Law (2010) sets general principles on renewable energy development.  See  Renewable Energy Law of the People’s Republic of China, Ministry of Commerce, People’s Republic of China. Available at: http://english.mofcom.gov.cn/article/policyrelease/ Businessregulations/201312/20131200432160.shtml Regarding the regulatory incentives for renewables dispatch, please see Chapter 4, especially Article 14 to 18. Also the NDRC issued notices on feed-in-tariff for polar energy (2011), geothermal power (2013) and wind power (2014). 36  See Chen et al. (2019) and Wei et al. (2018). 37  A 6% saving of total coal use as a result of fuller utilisation of more efficient coal-fired power plants would reduce CO2 emissions by roughly 248 m tonnes (or 0.5% of global CO2e). This is because coal consumption in the power and heat sector in 2014 was 1485 m tonnes (IEA (2016) 35

2  Lessons for China from International Experience of Power… 

39

savings of 6% are 1.7% of the value of total industrial electricity expenditure38; however, the price savings would be higher if a fall in coal marginal cost meant generally lower prices. Not all of the theoretical savings are realisable once genuine transmission and system stability constraints are taken into account and the fact that many thermally inefficient combined heat and power plants (CHPs) must run because of their heating loads in winter. CHPs accounted for 19% of the total power capacity in 2012 (220 GW) (CEC 2013). The most important point is that merit order dispatch underpins competition between power plants by massively sharpening incentives to cut running costs at individual plants in order to improve the probability of being dispatched. In addition, there is substantial inefficiency in regional power flows. This would be significantly helped by elevating the role of regional dispatch centres on the basis of least cost, as opposed to provincial dispatch, which favours provincial generation. A current constraint in reforming dispatch is the necessary software to optimise the operation of the system. The current estimates are that for most provinces or regions it would take 18 months to develop, test and implement new dispatch software (indeed the leading reform province of Zhejiang has taken rather longer than this to implement and test such software in a process which began in 2017). This does not seem a large medium-term barrier given that some provinces have now (mid 2019) got the necessary software in place. The bigger barrier is the impact on existing generator contracts, which are based on the expectation of sharing the available operating hours. For individual plants the financial impact of reallocating operating hours away from them would be substantial (e.g. consider two plants operating for 4000  hours on the basis of sharing hours, resulting in one operating for 7000 hours and the other operating for 1000 hours). The impact on the large generators might not be substantial at the corporate level given that income would simply be reallocated between plants. The fact that most of the power plants are in some Coal Information, Paris: OECD)), and 2.78 tonnes of CO2 are produced when 1 tonne of coal is burnt (i.e. 0.06∗1485∗2.78=248). 38  In all 6% of 1485 m tonnes of coal at 534 RMB/tonne divided by $402 bn of industrial electricity expenditure (2014 figures). A total of $402 bn is obtained by multiplying the industrial price figure of $0.1068 per kWh in Table 1.4 by industrial sales of 3770 TWh in 2014.

40 

M. G. Pollitt et al.

form of state ownership (though divided between national, provincial and local governments and their corporate investment companies) should facilitate internal public sector reorganisations of asset valuations. However, some compensation would doubtless be necessary at the generation company level and/or between different branches of government.

2.2.3 Privatisation and Monopolies (4) 2.2.3.1  Theoretical Significance Monopoly state ownership gives rise to a number of theoretical problems for the performance of an electricity sector (or any other state-controlled sector).39 These include arbitrary state interference in the operating and investment decisions of the sector; a lack of comparative information on performance of a given state-owned firm which allows its own managers and controlling ministry to evaluate and incentivise its performance; a lack of comparative information on the performance of different firms, which allows external regulators or financial investors to evaluate and incentivise performance; a lack of clearly defined or empirically justifiable corporate objectives, in contrast to a profit-driven private firm; lock-in to other forms of monopoly control, such as state control of hiring of senior managers, access to capital, control of input purchasing decisions; and exemption from or limitation of the rule of law towards the state-owned monopoly, leading to anti-competitive behaviour, lax health and safety regulation and environmental regulation and weak enforcement of rules and regulations (in contrast to private sector firms). Privatisation of a monopoly, even without any change in the structure of the firm, immediately exposes the firm to competitive forces and external regulation in the capital market, labour market and input market. It also reduces the scope for corruption and arbitrary sate interference. It generally sets up a longer run dynamic which will force a monopolist to improve its performance and be subject to pressure for further break-up under pressure from competitors and anti-trust authorities. The experience following the  For a good general discussion of privatisation, see Newbery (2002).

39

2  Lessons for China from International Experience of Power… 

41

1986 privatisation of British Gas as an un-restructured monopoly gas shipper-transmitter-distributor-retailer in Great Britain is an excellent example of this dynamic at work. The end point, in this case, was one of the most competitive wholesale (and domestic) gas markets in Europe.40

2.2.3.2  General Reform Experience The move away from monopoly state-owned enterprises in generation and transmission (e.g. CEGB in Great Britain, EdF in France, UES in Russia or SEGBA in Argentina) has generally been accompanied by significant privatisation of existing state-owned assets. The privatisation has often been consequent to the break-up of state-owned assets into separate competing companies for generation and monopoly transmission companies. In addition, there has been significant privatisation of distribution and retail companies (which were often separately constituted as municipal companies). There are well-documented cases of successful distribution of privatisations in the UK, Brazil, Chile, Peru and Argentina.41 Some jurisdictions have opted for partial privatisation. EdF remains only partly privatised and German utilities still have significant government ownership of their stock. In the Netherlands the distribution companies remain largely government owned, while they have all sold their retail businesses.

2.2.3.3  Chinese Context China has had a very significant reform of its electricity sector since 1985, but state ownership remains pervasive throughout the whole supply chain in electricity.42 There has been substantial entry of private and competing government-owned companies into the electricity generation segment. However much of this entry is directed by provincial investment  Contrast the initial negative assessment of Bishop et al. (1994) with an assessment of the impact of the privatisation in Florio (2004). 41  For the UK: Domah and Pollitt (2001); from Brazil: Mota (2003); from Chile: Pollitt (2005); from Peru: Anaya (2010) and from Argentina: Pollitt (2008a). 42  See Zhang and Heller (2007) and CNESA (2015). 40

42 

M. G. Pollitt et al.

companies who themselves are pursuing non-profit objectives. Most importantly seven state-owned companies—the Big five generators, State Grid and China Southern Grid—constitute around 50% of the total generation (see Fig. 2.3) and 100% of transmission, distribution and retailing.43 Private ownership is still quite limited in scope and there remain restrictions on private companies entering the generation market in particular. Privatisation remains a controversial pillar of the global electricity reform experience. Some studies have questioned whether it is a significant contributor to the improvement in performance of sector which liberalisation seeks to unlock.44 However, in the context of a large middle-­ income developing country, such as China, monopoly state ownership has a particular constitution and represents a significant, but subtle, barrier to improved performance. As was evident in the history of the state-owned generation and transmission monopoly in Great Britain (CEGB), state ownership of the electricity sector allows non-merit-based appointments of senior managers, limits the scope of competition in the input market and shapes investment decisions (often disastrously).45 It gives rise to significant scope for corruption and may put a break on competition and responses to price signals even where ownership is dispersed between different state-owned firms. While the European single electricity market is characterised by much continuing partial state ownership of generation assets, this does not impede the operation of a competitive market and severely limits the scope for non-market-driven investment decisions. European utility reform is also filled with examples of state-owned monopolies for which privatisation is the only reasonable solution to improve their performance (e.g. Network Rail and the Royal Mail in the UK46), in the face of the need to impose clear regulatory incentives on the access to capital for investment.  The subsidiaries of the ‘Big five’ are not included in the others in this graph, and the others include the private companies, international companies and other state-owned companies except the ‘Big five’. 44  See Pollitt (2005). 45  For a fascinating account of how this played out in the CEGB from 1962 to 1989, see Kim (2016). 46  See http://news.bbc.co.uk/1/hi/business/7401722.stm 43

2  Lessons for China from International Experience of Power… 

43

Installed capacity share 100% Others 80%

54.65% 51.68% 50.79% 51.62% 51.97% 53.66% 55.02%

60%

40%

China Guodian Corporation 6.73%

7.32%

7.23%

6.98%

7.13%

9.38%

9.86%

10.04% 10.47%

7.06%

9.76%

8.81%

9.14%

9.33%

8.86%

8.88%

8.97%

8.94%

10.40% 11.46% 10.96% 10.45%

9.92%

9.17%

8.79%

6.56% 8.86%

20%

0%

State power investment corporation

8.71%

10.83% 11.94% 11.74% 11.80% 11.78% 11.31% 11.06%

2008

2009

2010

2011

2012

2013

China Huadian Corporation China Datang Corporation China Huaneng Corporation

2014

Power generation share 100% 80%

Others 50.73% 50.67% 52.14% 53.36% 55.54% 56.24% 53.34%

60% 40% 20% 0%

State power investment corporation China Guodian Corporation

6.95%

6.89%

7.01%

10.08%

6.85%

9.59%

9.93%

9.82%

6.70%

9.93%

8.23%

8.49%

8.84%

8.83%

8.67%

8.59%

8.82%

11.18% 10.74% 10.15% 10.23% 10.59%

9.20%

8.75%

5.94% 8.63%

6.83%

8.40%

10.56% 11.41% 12.72% 12.78% 12.22% 12.09% 11.37%

2008

2009

2010

2011

2012

2013

China Huadian Corporation China Datang Corporation China Huaneng Corporation

2014

Fig. 2.3  Share of Big five generators in total capacity and total generation. (Source: Annual reports of power generation companies (2015))

44 

M. G. Pollitt et al.

Part-privatisation of the largest Chinese generating companies, or the wholesale privatisation of at least one of the Big five, would be a first step to significantly reducing the monopoly control of the central Chinese government and allowing the government to experiment with the benefits of loosening state control on a potentially competitive part of the electricity sector.

2.3 Supportive Secondary Market Arrangements 2.3.1 C  reation of Spot and Ancillary Services Markets to Support Real-Time Balancing of the System (5) 2.3.1.1  Theoretical Significance A market-based electricity system involves using the pricing mechanism to price the different electricity products that are necessary to supply kWhs to final customers at the right quality (see Stoft 2002, for a detailed discussion). A full set of markets would include spot and forward markets for real energy, and markets for frequency response, reactive power, voltage regulation and reserve capacity (see Fig. 2.4). Spot markets reward generators for matching supply and demand in near real time. This typically involves day-ahead markets and intra-day (balancing markets). This gives incentives to generators (and to loads) to adjust their position on the basis of the general condition of the system and their own operating situation. Longer-term markets—such as monthly or yearly contract markets—offer financial hedging to generators and loads. Ancillary services are to do with maintaining power quality in real time. Traditionally these have been quite a small part of the electricity market in systems characterised by large fossil fuel-based power plants. This is because these types of plants can cheaply provide ancillary services

2  Lessons for China from International Experience of Power… 

45

Fig. 2.4  List of electricity product markets, including ancillary services products. (Source: Stoft (2002, pp. 82, 236))

as by-products in the production of real energy. As power systems become more complex and involve more renewables the market efficiency of ancillary services provision becomes more important. In Great Britain ancillary services are expected to grow from 2% of wholesale electricity costs in 2015 to 25% by 2030.47

 See Newbery et al. (2016) for a discussion of the possible increase in importance of ‘flexibility’ markets within the electricity system by 2030 in Great Britain. 47

46 

M. G. Pollitt et al.

2.3.1.2  General Reform Experience The development of spot markets for electricity has been central to the development of wholesale power markets. A spot market provides the basic price signal around which all futures prices can be determined. This is true in all energy-based commodity markets. This is because spot prices can provide transparent constantly updated information and are the basis for the determination of longer-term contract prices. Efficient spot markets are visible to entrants and can give good signals on the viability of entry at any given moment. They are also extremely important for signalling to the demand side of the market the value of short-run actions to reduce (or increase) power demand. Sophisticated electricity markets around the world have been developing their markets for ancillary services.48 This has been because a sharpening of the incentives around the delivery of real energy means that power quality must be appropriately rewarded or there will be pressure for it to deteriorate. This is most obvious in the area of reserve capacity where the move towards a liberalised market means a (necessary) reduction in the holding of reserve capacity. If this leads to the system operating at an unacceptably high risk of a rolling blackout there may be a case for creating a market to specifically reward capacity (separately from energy). However, it is fair to say that incentives for rewarding ancillary services remain a patchwork of no-payment (compulsory provision), fixed payments, bilateral contracts and bid-based markets.

2.3.1.3  Chinese Context Individual power plants in general receive a regulated payment for the power that they generate. This is a negotiated price that varies at the provincial level and agrees with the local National Development and Reform Commission (NDRC) on the basis of local production costs and socio-­ economic conditions. The aim is to allow a plant to recover a reasonable rate of return considering its costs and the number of hours it is likely to  See Pollitt and Anaya (2016) for a discussion of recent market developments in Germany, Great Britain and New York. 48

2  Lessons for China from International Experience of Power… 

47

run for (see Ma 2011). Plants are allocated hours on the basis of ‘equal share dispatch’ (see Karhl et al. 2013, 2016) discussed earlier. The current terms appear to be very generous (see Table 2.1) and encourage the building of new generation capacity (see Rioux et al. 2016). China has a coal-based power generation system that has minimised the need for formal ancillary services markets.49 There are some payments to generators that must run for system support (voltage support or reactive power) reasons; however, there is, in general, no formal payment mechanism for ancillary services. There have been suggestions of how the market might be reformed in the Chinese context.50 If power dispatch were to be reformed in China, the lack of formal mechanisms for procuring ancillary services would be more of a problem because some plants which are needed for ancillary services might be in danger of closing on the basis of their lack of competitiveness in the wholesale energy market. Hence one can envisage that reform of dispatch would require reform of ancillary services payments.

2.3.2 P  articipation of Demand Side in Wholesale Electricity Markets (6) 2.3.2.1  Theoretical Significance One of the cheapest sources of power available in any electricity market is demand reduction. This is where some demand that would otherwise be on the system is paid to not run. This is a key source of flexibility in any electricity system and particularly useful in managing peak demand and the requirements for reserve capacity. Unlocking demand reduction is a low-cost way of increasing competition in the market and balancing supply and demand. The introduction of a spot market for electricity greatly facilitates demand-side participation, because this is the market in which demand can most easily participate. Demand-side participants are frequently large industrial users who can turn down or reschedule their 49 50

 See Ming et al. (2014).  See Zheng and Zhou (2003), Ming et al. (2014), Yao et al. (2015) and Mingtao et al. (2015).

48 

M. G. Pollitt et al.

production in response to system conditions. Industrial users can sign contracts with aggregators that pay them significant amounts per kWh of demand they reduce at peak times (or have signed contracts which give them otherwise cheap power, but then expose them to very high prices if they consume it at times of annual system peak demand). With the emergence of distributed electrical energy storage the capacity for industrial and commercial loads to further manage their interaction with the wholesale market will increase.

2.3.2.2  General Reform Experience Demand-side participation has been very significant in competitive power markets (see Table 2.2)51 in that there have been many occasions where there would been blackouts if there had not been active demand-­ side participation in wholesale power markets (in Great Britain there have been several incidents of demand-side participation helping the system cope with extreme system conditions). Demand-side participation has been very important in ancillary services markets. Most recently capacity markets in both the US and Great Britain have seen significant price-reducing impacts from the inclusion of the demand side in the market.

2.3.2.3  Chinese Context One of the limitations on demand participation is that domestic and high-value commercial loads are more expensive to incentivise to reduce or shift their demand (per kWh of demand reduction). The current power Table 2.2  Demand participation in liberalised markets (DSR = demand side reduction)

DSR as % of peak

Texas ERCOT

Great Britain National Grid

US PJM

3.20%

3.60%

9.10%

Source: Khalid (2016, p. 3)

 See Taylor et al. (2014).

51

2  Lessons for China from International Experience of Power… 

49

sector reform is envisaged to increase the role for demand-side response.52 This implies that de-industrialising economies such as in the US and Great Britain have smaller, easier-to-shift industrial loads over time. This is not the current situation of China, as shown in Fig. 2.5, where 71% of power demand is from industry (as opposed to only 26% in the US). This suggests a high potential for demand-side response in China.53 Projections from Shanghai and the actual experience of Jiangsu suggest a significant and rising percentage of peak demand response as shown in Table 2.3. In Jiangsu province, the grid companies and related governmental agencies launched a demand-side management pilot project in the summer of 2016. In all 3154 households participated in this DSR pilot project that saw a peak demand reduction of 3.8% (the equivalent of 3520 MW).54 It is important to point out that demand response depends on customers being subjected to sophisticated metering which can measure their consumption, when they are meant to be offering demand response. In Yunnan, in 2015, only half of all industrial customers had got meters capable of allowing them to participate in a spot market. It also required that the demands themselves are subject to market-based incentives for their output otherwise they may have an incentive to game their participation in demand-side response by deliberately running their equipment ahead of an instruction to reduce their demand.

2.3.3 R  egulated Third-Party Access to, and Efficient Allocation of, Scarce Transmission Capacity (7) 2.3.3.1  Theoretical Significance Transmission capacity is a scarce resource.55 This is because transmission system investments are large and eventually it becomes difficult to expand  See Lei et al. (2018).  See NDRC (2010), Crossley (2014) See also Zhang et al. (2017) and Wang et al. (2010). 54  The data of Jiangsu province are from http://www.sdpc.gov.cn/fzgggz/jjyx/dzxqcgl/201607/ t20160727_812571.html 55  See Hogan (1992). 52 53

50 

M. G. Pollitt et al.

Power demand share in China in 2014 (TWh) 106 , 2% 101 , 2%

72 , 1%

200 , 4% Industry 361 , 7% Residential Consumption 718 , 13%

Others

3770 , 71%

Wholesale, Retail Trade and Hotel, Restaurants Transport, Storage and Post

Power demand share in US in 2014 (TWh)

8.87, 0.24%

958.91, 25.71%

1,359.54, 36.46%

Residential 1,401.67, 37.59%

Commercial Industrial Transportation

Fig. 2.5  Sources of electricity demand in China and the US. (Source: NBS (2015) and EIA (2015))

2  Lessons for China from International Experience of Power… 

51

Table 2.3 The potential for demand response in Shanghai and Jiangsu in selected years Jiangsu-2016 Shanghai-2020 Shanghai-2025 Shanghai-2030 DSR as % of peak

3.8%

1.68%

3.04%

4.09%

Source: The data on Shanghai are from Liu et al. (2015, p. 16), and the data of Jiangsu are from http://www.sdpc.gov.cn/fzgggz/jjyx/dzxqcgl/201607/ t20160727_812571.html

transmission capacity as a result of objections from the communities through which they pass (and who do not benefit directly from them). Because of loop flows in meshed electricity networks and constraints within the distribution system to which they are connected, transmission building to solve one constraint may create other constraints and vulnerabilities. Building appropriate transmission capacity in the face of rapid growth of generation or loads in some areas (rather than others) often gives rise to transmission constraints in developing countries. Transmission capacity can often only be added in significant-sized increments, meaning that until the next incremental expansion of capacity comes along there will necessarily be constraints within the transmission system. Every advanced country in the world has had difficulty in expanding transmission capacity beyond a certain point. This suggests that a system for allocating the available transmission capacity is necessary. This is important when the cheapest generation sources are a long way from the largest and most valuable loads. Allocating transmission capacity among generators and loads in an efficient way and providing signals for where to expand the transmission system next are important elements of a reformed electricity system. The allocation of capacity among generators and loads should be non-discriminatory. This is important in countries where transmission capacity is still owned by an incumbent generator, to prevent unfair access terms reducing competition in the wholesale market. One way to indicate where the constraints are in the transmission system is to use short-run price signals at every node in the transmission system—locational marginal prices (LMPs)—which indicates a different price for wholesale power between export (lower) and import (higher)

52 

M. G. Pollitt et al.

constrained nodes.56 This provides signals to switch off high-cost generators and switch on  lower value loads in export-constrained areas and switch off lower value loads and switch on higher-cost generators in import-constrained areas. Those using the transmission system have to pay the differences between LMPs at each end of a line to use the system in the direction that the power is flowing. Another way to do this is to allocate transmission system costs by location such that generators connecting in export-constrained areas face higher use of system charges than generators connecting in import-­ constrained areas and vice versa for loads.57 Both of these methods help make good use of the existing transmission system and signal the value of transmission system expansion on the basis of the modelled reduction in constraint-related payments.

2.3.3.2  General Reform Experience Non-discriminatory access to the transmission system based on regulated third-party access has been a central element of power market reform as countries have moved away from integrated generation and transmission utilities. There are well-documented cases where continuing integration of generation and transmission did lead to preferential access being given to incumbent generators by their transmission business (e.g. in Chile). This resulted in long-running competition disputes.58 The EU stated that ownership unbundling of transmission from generation with regulated third-party access (a single set of access prices) was its preferred model for the organisation of the transmission system. Allocation of transmission capacity has been largely done on the basis of published tariffs with firm transmission rights (guaranteed access). This results in compensation being paid in the event that generators or loads have to reduce their supply or demand due to transmission  See Bohn et al. (1984).  This is the case in Great Britain, for instance, where there is a substantial difference in transmission charges for loads in Scotland (lower) and in London (higher) due to transmission constraints between generation centres in the north and load centres in the south. 58  See Pollitt (2005). 56 57

2  Lessons for China from International Experience of Power… 

53

constraints. In Great Britain such firm transmission rights do come with the locational signals in the use of system charges for connecting generation in demand-constrained parts of the network and vice versa. Independent system operators (ISOs) in the US mainly use LMPs to optimise the use of the transmission system in real time and then allow the trading of financial transmission rights (FTRs)59 allocated to incumbents every six months that allow transmission system users to hedge their exposure to LMPs. For significant transmission links, such as long distance High Voltage Direct Current (HVDC) lines in the US60or international transmission links in Europe,61 capacity is auctioned and this is a way of efficiently allocating the capacity among users. However, such congestion-based mechanisms for charging do not guarantee that the fixed costs of the line will be recovered and hence additional charges will be almost certainly be needed.

2.3.3.3  Chinese Context Recently China has been rapidly building generation capacity, expanding power demand and building new transmission lines. It has invested significantly in HVDC links.62 The rapid expansion of the high-voltage grid can be seen in Table 2.4. The allocation of the capacity within the transmission system in China is rather underdeveloped. Thus some significant capacity remains underutilised (e.g. between Yunnan and Guangdong) and other lines are not being allocated efficiently (e.g. to renewables from low cost regions). This suggests that some reform of the pricing and allocation mechanisms around transmission capacity would be beneficial and would be in line with other reforms to dispatch and wholesale power markets. Under the current system of charging final customers transmission and distribution charges have not be separately identifiable, even for the largest customers. This is because both final customer and generation prices  Chao and Peck (1996).  See Archer et al. (2017). See also Zhou et al. (2016). 61  See Bergman et al. (1998). 62  See Zheng et al. (2016). 59 60

54 

M. G. Pollitt et al.

Table 2.4  Expansion of Chinese transmission and distribution grid Type

Voltage 2017

AC 1000 kV 10,073 (km) 750 kV 18,830 500 kV 173,772 330 kV 30,183 220 kV 415,311 110 kV 631,361 35 kV 508,682 Total 1,788,212 DC 800KV 20,874 (km) 660KV 1334 500KV 13,552 400KV 1640 Total 37,399

2016

2015

2014

2013

2012

7425 17,968 165,875 28,366 397,050 611,431 499,400 1,727,333 12,295 1334 13,539 1640 28,808

3114 15,665 157,974 26,811 380,121 591,637 496,098 1,671,420 10,580 1336 11,872 1640 25,429

3111 13,881 152,107 25,146 358,377 566,571 484,296 1,603,489 10,132 1336 11,875 1640 24,983

1936 12,666 146,166 24,065 339,075 545,815 464,525 1,534,248 6904 1400 10,653 1031 19,988

639 10,088 137,104 22,701 318,217 517,983 456,168 1,462,900 5314 1400 9145 1031 16,890

Source: China Electricity Council (2018)

have been regulated by the government. Unit charges have varied for different customers connected at different voltage levels (as would be suggested by optimal allocation of fixed costs between customers). Recently the government has moved away from per unit charging to introducing fixed fees for some large industrial customers.63 Under the recent market trials transmission/distribution network charges at the provincial level have been identified. However these new transmission and distribution charges in China are cost plus and do not incorporate location or time of use signals of the longer run or real-time condition of the network. Generators do not bear any of the costs of transmission system (unlike in some systems such as the UK) and rely on curtailment and regulated energy prices to signal advantageous locations for connection. Reforming the charging for transmission to better allocate the available capacity would seem to be advantageous especially in signalling where to locate renewables, build new transmission capacity and which generation plants should run in constrained areas of the network.

  See National Development and Reform Commission, Available at http://www.sdpc.gov. cn/zcfb/zcfbtz/201607/t20160706_810665.html and http://www.chinasmartgrid.com.cn/ news/20160909/618793.shtml 63

2  Lessons for China from International Experience of Power… 

55

2.4 Appropriate Economic Regulation 2.4.1 U  nbundling of Regulated Network Charges and Competitive Segment Charges (8) 2.4.1.1  Theoretical Significance It was Joskow and Schmalensee (1983) who definitively pointed out (following Weiss 1975) that wholesale generation and retail service were potentially competitive elements within the largely vertically integrated US electricity power industry. Only transmission and distribution networks—which provide the transport capacity to the electricity system—have the characteristics of natural monopolies. Even then distribution networks are local monopolies and multiple distribution systems can coexist within a single country providing the ability to regulators to benchmark these local natural monopolies against one another using yardstick competition (following Shleifer 1985). Transmission monopolies can exploit economies of scale over larger areas and over individual lines, but the wide area natural monopoly is actually in system operation rather than the ownership and operation of the lines themselves. Thus in many parts of the US transmission ownership is actually dispersed among local companies, while system operation (such as in PJM) is conducted over a wide area. Such a separation between transmission and distribution networks and the rest of the system allows pricing mechanisms to be clearly distinguished. Prices for wholesale and final retail charges can be competitively determined, while both the level and structure of distribution and transmission charges continue to be regulated. Attention can be paid to ensuring that such charges are non-discriminatory, that is they do not favour any particular user of the network on the basis of their ownership characteristics, in particular whether they are part of the same company that owns the transmission and distribution system.

56 

M. G. Pollitt et al.

2.4.1.2  General Reform Experience The increasingly strict unbundling of transmission and distribution charges from charges for wholesale and retail elements of electricity supply has been a key element of both the successful creation of competitive wholesale and retail electricity markets around the world. In the UK transmission and distribution network ownership was unbundled from generation ownership at the time of privatisation. At the level of the EU successive electricity directives (1996, 1999 and 2003) required accounting and legal unbundling of the network elements from the rest of the system.64 This meant that transmission and distribution businesses must be created within companies that remain integrated with generation and retail. Transmission and distribution must be legally unbundled from each other within the EU.  The EU has expressed a preference for the ownership unbundling of transmission from the rest of the electricity system, given the key role that non-discriminatory access to the transmission plays in promoting retail competition. As we discussed earlier, strict separation promotes a level playing field in the competitive segments of the electricity supply industry. A significant share of the total benefit of liberalisation arises in the network businesses themselves. A key success of the separation of network and competitive elements in many countries has proved to be the ability to introduce incentive regulation of the network businesses. This has involved CPI-X regulation of the revenue of network companies, with formulae set in advance for several years (usually 3–5 years). This has resulted in very significant improvements in the efficiency of operation of the network companies. In the UK perhaps one-third of the overall gain from the liberalisation process came from improved regulation of the network companies, rather than competition per se (see Littlechild 2006 and Pollitt 2012).

 See Jamasb and Pollitt (2007).

64

2  Lessons for China from International Experience of Power… 

57

2.4.1.3  Chinese Context The power sector reform of 2002 resulted in a very significant set of measures aimed at separating networks from competitive elements. This did result in separation of generation from the rest of the power system. However, as we have already observed transmission, distribution and retail remain bundled within State Grid and China Southern Grid.65 Generation is under contract to supply the grid companies with power to supply their final customers. This is a form of single-buyer model, which was used in the early days of power market reform in some countries. It was an option under the 1996 electricity directive in the EU. This model has now been discontinued in the EU (for fossil fuel generation) in favour of competitive wholesale power markets to determine the price of bulk power. As part of the reform process, the provinces (including Guangdong) have announced network access charges which generators need to pay to use the transmission and distribution system to competitively sell power to final customers.66 These charges are based on the identification of a regulatory asset base for transmission and distribution assets within the province and the calculation of what charges would allow the relevant grid company to recover a fair return on this asset base while covering its costs. The initial charges that have been announced are fixed for three years. This would seem to give some incentives for the grid company to cut its network costs and keep the savings. The move towards stricter separation of network and competitive elements within the power system in China is to be encouraged. For China to bring itself in line with the international best practice, there must be strict accounting and legal separation of transmission, distribution and retail businesses. Such strict unbundling would allow the publication of data on network company costs and facilitate independent comparative benchmarking of regulated network businesses, which is currently very difficult to do on the basis of the high level horizontally and vertically integrated business data that are available on China State Grid and China Southern Grid at the moment. This will greatly facilitate 65 66

 See Li et al. (2016).  See Zheng et al. (2016) and Alva and Li (2018, p. 50).

58 

M. G. Pollitt et al.

non-­discriminatory access charging for the use of the distribution and transmission system. It will also allow incentive regulation of transmission and distribution to be introduced. An obvious way forward is to compare distribution and transmission costs at the provincial level and use benchmarking to compare costs and set the efficient level of revenue for the transmission and distribution elements. Setting a CPI-X price cap for three  years is a good start and does allow some differences in performance to emerge quickly, before moving to setting the price controls for longer periods (in the UK, this was initially five years for distribution and four years for transmission, later five years for both and now eight years for both). Regulation of investment is also important, given the high rates of investment in networks in recent years. There is currently a lack of incentives to limit these investments, in contrast to jurisdictions with incentive regulation where sophisticated audits and menu regulation have been developed to limit overinvestment by monopoly network companies.67

2.4.2 M  echanisms to Ensure Competitive Procurement of Wholesale Power for Regulated Final Customer Groups (9) 2.4.2.1  Theoretical Significance Unless all of the retail market is liberalised there will continue to be significant numbers of customers who are on regulated final tariffs. If this is the case these customers need to be supplied with wholesale power, which has been procured on a competitive basis. This is because if they are not, this will significantly reduce the degree of competition in the wholesale market. There is no reason why all retailers should not procure their power competitively in the wholesale market, whatever the basis of the final contracts that they need to offer to customers in the regulated retail price market.68  See Jamasb and Pollitt (2007) on the UK.  For a discussion of this in the context of Ohio, see Littlechild (2008).

67 68

2  Lessons for China from International Experience of Power… 

59

2.4.2.2  General Reform Experience Most countries across the world that have competitive wholesale power markets also have regulated final tariff customers. This is case in half of all EU countries (see ACER 2015), most of the US and all of South America. These customers are the default service contract customers in the US and the EU and most of the final household customers in South America who are still on regulated final tariffs. In all of these cases the procurement of the power to supply these protected customers is done on a competitive basis. This is achieved by the regulator specifying the basis of the contract for bulk power that it allows to be passed through to the regulated final customers. In Denmark, the regulator specifies a default mark-up formula for residential customers.69 This is based on regulated mark-up competitively acquired wholesale electricity.70 In the US the wholesale power contract to supply default service customers within a particular distribution company area is often auctioned and the auction price is then used to price the wholesale cost element of the default service bill that residential customers are charged.71

2.4.2.3  Chinese Context Currently, the two large grid companies procure power at regulated prices for all their customers.72 The final prices that they can charge and the prices that they pay for wholesale power are regulated. It is highly likely, and in line with international experience, that China will want significant numbers of customers to continue to enjoy regulated tariffs for the foreseeable future. This is particularly true in the residential sector where customers are currently paying below the full economic cost of their service.73 If Chinese residential prices were raised to US levels this would allow industrial prices

 See Danish Regulatory Authority (2014).  See Danish Regulatory Authority (2014). 71  See Littlechild (2008). 72  See Ma (2011). 73  See Feng (2016b). 69 70

60 

M. G. Pollitt et al.

to be reduced by up to 5%.74 Indeed, there are no plans to liberalise the market for residential customers in this round of reforms. This may be because of a debate among senior policy-makers about whether electricity is a commodity (which should be priced to reflect costs) or a public service (which should continue to be cross-subsidised).75 The continuing presence of default service customers does not mean that there needs to be anything less than full competition in the wholesale market for power. This can be achieved by moving to competitive procurement for wholesale power from fossil fuel power plants. Such a mechanism could also be used to introduce regular cost-based updating of retail prices on the basis of changes in underlying power procurement costs. This would be the basis for gradually raising retail prices, as incomes continue to grow, towards fully cost reflective levels. Clearly identifying the procurement costs associated with default service customers, combined with separate network charging, would also clearly identify the level of the subsidy that these customers are currently receiving. This would have the additional advantage of focussing regulatory attention on how this might be reduced over time.

2.4.3 T  he Creation of Independent Regulatory Agencies to Regulate Monopoly Network Charges and Monitor Competitive Segments (10) 2.4.3.1  Theoretical Significance Competitive wholesale and retail markets need to be monitored carefully to ensure that they are working properly. This is because they are  The US residential prices in 2014 were $0.125 / kWh, in China they were $0.0907 / kWh. Raising Chinese prices by 38% would raise an additional $14.65 bn of revenue (assuming a demand elasticity of 0.3 (He et al. 2011) and initial sales of 718 TWh). This would (roughly) allow for around a 5% drop in the industrial price of $0.10675/kWh, assuming an elasticity of 0.18 (He et al. 2011) for industrial demand (this can be roughly calculated by adding three effects: the initial additional residential revenue equals 3.6% of initial industrial revenue + industrial demand response from the price drop (a further 0.7% or 0.18 ∗ 3.6%) + system savings due to reduced residential demand (0.5% of total system costs due to aggregate demand falls if industrial prices fall 5% and residential prices rise 38% and the marginal system saving is the coal cost of $0.038/kWh)). 75  See Xu (2017, p. 117). 74

2  Lessons for China from International Experience of Power… 

61

creations from incumbent monopolies and exhibit natural tendencies to re-­integrate. This implies that it is unlikely that a general competition authority will be nimble enough to confront all of the many competition issues that are likely to arise, especially in the early years following liberalisation. In addition, substantial regulated monopolies remain in the sector. These need to be regulated as to the level and distribution of their charges and the quality of service that they are offering to both their retail and generation customers wishing to use their networks. Such regulation is a substantial task and requires detailed knowledge of the cost structures of the industry and attention to the incentive effects of any financial controls that are put in place. These two facts suggest that in line with other significant utility industries (telecoms, gas, rail and water) a dedicated regulatory body may be best placed to ensure society’s continuing interest in these sectors. The institutional form of such a regulatory body can be debated and depends to some extent on the size and competence of state in which it is situated. It could be merged with other regulated industries (such as in Germany within the Bundesnetzagentur), involve both national and subnational bodies (such as with Fel Electricity Regulatory Commission [FERC] and the state Public Utility Commissions [PUCs] in the US), be a division of the competition authority (as in the Netherlands with the DTe being merged into the NMa) or involve a separate electricity regulator (e.g. ANEEL in Brazil) or a combined electricity and gas supply regulator (e.g. Ofgem in Great Britain). In most countries with liberalised power markets, the regulatory body tasked with overseeing competition and monopoly regulation is independent of day-to-day central government control in the sense that it is a non-ministerial government department, where the relevant government ministry (of energy) has limited powers of intervention during the term of office of the board members of the regulatory body. Both the World Bank and the EU have strongly endorsed this approach to regulation.76 This is because a key problem in liberalised network industries is one of  See Bessant-Jones (2006) for a World Bank view on independent regulation and power market reform; and Jamasb and Pollitt (2007) and Pollitt (2008b) on independent regulation within the EU electricity directives. 76

62 

M. G. Pollitt et al.

regulatory appropriation (see Gilbert and Newbery 1994). This occurs when governments have the incentive to encourage private companies to invest and then to force them to reduce prices after investment has occurred in order to ‘appropriate’ a greater share of the benefits from investors to customers. ‘Independent’ regulation is primarily aimed at balancing the rights of shareholders to return with the rights of consumers to fair (i.e. reflective of competitive cost levels) prices.

2.4.3.2  General Reform Experience The experience of liberalised markets is that independent regulators have supported private investment in liberalised power markets, have had significant roles in monitoring day-to-day competition issues and have made significant progress in developing network regulation. The roles of Offer (the GB electricity regulator from 1990–99) and then Ofgem (the electricity and gas regulator from 1999) have been very significant in the Great Britain context. The presence of a regulatory body with an appointed regulator (and then regulatory board) with statutory duties (in particular, to promote competition) gave investors confidence that the government would not arbitrarily intervene to reduce prices. This resulted in significant new investment in the industry in the years following privatisation and eventually significant foreign investment in the sector (which saw the assets being sold at a large premium to overseas investors). It also resulted in close regulatory oversight of the process of competition in the early years (see Newbery 2005) that did eventually result in competition authority enforcement action (sanctioned by Ofgem) to further break up the incumbent generators. Offer and Ofgem developed very sophisticated and successful incentive regulation of network companies which saw the level of real charges fall nearly 60% in distribution and 40% in transmission between 1990 and 2005 (see Jamasb and Pollitt 2007, and Ofgem 2009). These results have been mirrored elsewhere and Cubbin and Stern (2006) found significant investment benefits in electricity resulting from privatisation and independent regulation across a sample of countries, many of which had previously suffered from chronic under-investment in electricity infrastructure.

2  Lessons for China from International Experience of Power… 

63

A particular success of the regulator in the UK has been to counteract the power of incumbent companies in the sector. This was greatly facilitated by the break-up of the incumbent generator and the separation of the transmission network company from the rest of the system. The regulator has been a consistent advocate for introducing more competition, for instance in the procurement of network assets, and for changes to network and industry rules which increase costs for customers. The regulator has also been a significant source of learning in the sector as problems have been revealed and dealt with and new issues have come to light as the reform has progressed. Many developing countries have set up nominally independent regulatory agencies for electricity. These often suffer from a lack of genuine political will to leave the sector to be overseen by the regulator and a lack of resources on the part of the regulatory agency to effectively enforce competition and network regulation (see Pollitt and Stern 2011). Regulatory agencies in many countries need a combination of well-­ trained economists, lawyers and accountants to adequately undertake economic regulation. Low civil service pay in competition with relatively well-resourced incumbent companies makes it difficult to attract  and retain high-quality staff with relevant industry knowledge and experience to work in regulatory agencies in many countries.

2.4.3.3  Chinese Context As suggested by Fig. 2.6, Chinese regulatory oversight of the electricity is complicated.77 There was an attempt to create a separate economic regulator in 2003 within the NDRC (the State  Electricity Regulatory Commission) to oversee competition and pricing, but this was  later merged back into the National Energy Administration (NEA). Determination of regulated final prices and the prices paid to generators is currently split between the provincial and national Pricing Departments of the NDRC, which is the government department that oversees economic reform across the whole economy. None of the bodies (in Fig. 2.6) 77

 See An et al. (2015).

64 

M. G. Pollitt et al.

Fig. 2.6  Current structure of regulatory bodies overseeing the Chinese electricity sector. (Notes: Ministry of Environmental Protection [MEP], National Development and Reform Council [NDRC], State Administration of Work Safety [SAWS], National Energy Association [NEA], State-owned Assets Supervision and Administration Commission [SASAC]. Source: Tan and Zhao (2016))

has exclusive power over coordinating the electricity policies and none want to be coordinated by others. Therefore, it is sometimes difficult to balance the actions of different government agencies. One consequence of this is that the energy policy for the 12th Five-Year Plan (2011–15) did not come out until 2013.78 China does not have a tradition of independent regulation and even in the telecoms sector (where, internationally, deregulation is normally more advanced than energy) there is no regulator separate from the Ministry of Communications.79 The situation is complicated by the continuing state ownership of most of the electricity supply sector and the  See Xu (2017, pp. 83, 122, 126).  See Yeo (2008).

78 79

2  Lessons for China from International Experience of Power… 

65

role of the State Asset Holding Company.80 However there has been some success in recent years in improving the functioning of the general competition authority in China, which has become more active in monitoring and enforcing competition across the economy.81 Section 7, Paragraph 2 of China’s Anti-monopoly Law (at the start of 2015) does cover state-­ owned enterprises (SOEs) and prohibits the abuse of dominant positions. However, the law also protects SOEs that ‘implicate national economic vitality and national security’ and hence there is a limitation to the extent to which current anti-monopoly legislation covers large SOEs in the electricity sector. Anti-monopoly enforcement activities in China in 2015 were split between three branches of government (see Fig. 2.7). However

Fig. 2.7  Anti-monopoly institutions in China in 2016. (Source: Slaughter and May (2016, p. 2))

80 81

 See China National Energy Administration (2016, 2015)  See Slaughter and May (2016).

66 

M. G. Pollitt et al.

Table 2.5  Evidence on civil service pay in China relative to state-owned companies 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017

Electricity, gas and water utilities (1)

Public sector (2)

Ratio (1)/(2)

121 140 161 185 218 252 276 312 349 383 445 490 523 549 597

100 113 132 147 181 211 231 250 275 301 322 347 407 463 525

1.21 1.24 1.22 1.26 1.20 1.19 1.19 1.25 1.27 1.27 1.38 1.41 1.29 1.19 1.14

Note: Public sector in 2003 = 100 Source: NBS, Statistical Yearbook of China, various years

there is evidence that the Anti-monopoly Enforcement capability of the Chinese government is increasing with a new single anti-monopoly body—State Administration for Market Regulation (SAMR)—being introduced in 2018.82 The development of the powers of this new body is important if the role of the market is extended in electricity, as it has been in advanced countries.83 Regulators can only be as effective as the quality (and quantity) of the staff that they have. Civil service pay remains relatively low in China and this is a problem in recruiting and retaining staff to undertake regulatory functions. There is evidence that salaries in government remain low relative to those in the SOEs that they regulate (see Table 2.5), though the gap is smaller than it was. A well-resourced economic regulator is an essential part of a successful ongoing reform of the electricity sector in China. FERC in the US has  See https://www.chinalawvision.com/2019/02/uncategorized/new-era-comes-highlights-antimonopoly-law-china-2018/ 83  In the UK, arguably the competition authority has been more significant in advancing competition in the wholesale electricity market than the regulator (see Newbery 2002). 82

2  Lessons for China from International Experience of Power… 

67

1500 employees (working on electricity and gas) working at the fel level, with significant additional numbers in the State PUCs. Ofgem in Great Britain has 907 employees (for a population of 58 m). How to ensure the independence of this regulatory agency in a Chinese context is challenging, given the lack of experience in China with independent regulation. One solution is to incorporate the electricity regulator into the general competition authority to begin with. This would have the advantage of both strengthening functions of the authority and allowing for the initial focus to be on promoting competition (which is the key thrust of the No.9 document). Another way forward would be to create an independent regulator with board members appointed for five years and use this as a test case for reforming utility sectors in China. These board members should be from a mixture of backgrounds and consist of both executive and non-­executive members of the regulatory agency. The price from getting incentive regulation of the electricity distribution network in China is large. Consider Fig. 2.8. In Great Britain, between liberalisation in 1990 and 1998 employment dropped 43% and labour productivity in electricity transmission and distribution improved over 100%. Given that China State Grid and China Southern Grid have 2  m employees, the savings in unit labour costs are significant (conservatively), estimated at $8.7 bn p.a. or 2.1% of industrial customer expenditure; thus, prices could be reduced by the order of 2–3%.84 Not all of the total number of employees are involved in the electricity supply business—a significant number are in other activities, potentially reducing the scope for rationalisation.85 According to an NEA regulation report, there is also a problem around grid asset depreciation in China’s grid accounting, which can be seen as a strategy to inflate current costs (and charges). For example, the official  In all 43% of the 2015 employees (870,000) could be released from the sector at $10,000 per employee. Industrial demand is 3770 TWh, paying $0.1068 per kWh. The average wage in China was 62,029 RMB or $8965 (6.92 RMB to the $). Industrial price elasticity is assumed to be 0.018. 85  For example, State Grid reportedly had 1.5 million employees in total in 2005, of which 72% were in the electricity business (generation, transmission and retailing), 21.6% were in construction and 2% were in research and design (see Xu 2017, p. 142). 84

68 

M. G. Pollitt et al.

GWh distributed per employee

16 14 12 10 8 6 4 2

es 015 ) ur ( rg en Ede 199 3 tin s a_ ur ( ) A rg 2 E 00 d en 6) Br tin eno az r a (1 _E il( 99 19 d 3) Br 91 eno az r ) _W (2 il( 00 20 ho 6) 00 le Pe ) c ru _W ou (1 nt ho ry l Pe 986 )_ e co ru Si u (2 nt x 00 r 7) com y _S pa Ch ix ni es c ile _C omp hi an Ch le ie ct s En ile_ ra C ( gl 1 h 9 an ile 87 d ) an ctra En ( d gl W 200 an a 3) d an les (1 d 9 W 9 al es 0) (1 99 8)

(2

CC

a_

A

A

rg e

nt

in

a_

Ed

G

in

_S

A

rg e

nt

na

Ch i

Ch i

na

_C SG

(2

01

5)

0

Fig. 2.8  International comparison of impact of reform on labour productivity in electricity distribution and transmission. (Notes: The six companies from Peru in this figure are Electro Sur Medio, Electrolima, Edelnor, Luz del Sur, Ede Chancay and Ede Cañete. CSG = China Southern Grid; SGCC = State Grid Corporation of China. Sources: Anaya (2010), China Electricity Council (2015), Domah and Pollitt (2001), Mota (2003), National Grid Electricity Transmission Report and Accounts, Pollitt (2004 & 2008))

depreciation period of one transmission line belonging to Guizhou grid is 17 years; however, the actual depreciation period is less than 5 years.86 Transmission assets are typically depreciated over 40 years or more in EU countries. Rapid depreciation of new assets in an expanding system raises measured costs and hence prices. Proper regulation of grid asset accounting and their translation into electricity prices might produce further significant savings for industrial electricity customers.

 See http://zfxxgk.nea.gov.cn/auto92/201606/t20160614_2264.htm

86

2  Lessons for China from International Experience of Power… 

69

2.5 E  fficient Promotion of Low Emission Technologies 2.5.1 C  ompetitive Procurement Processes for Low Carbon Generation, with Some Exposure to Wholesale Price Variability (11) 2.5.1.1  Theoretical Significance Low carbon generation is, mostly, not currently financially cost competitive with electricity produced from fossil fuels. This implies that if governments want to support low carbon generation they need to find ways to implicitly (e.g. by banning fossil fuel use) or explicitly (e.g. by making use of feed-in tariffs) subsidise it. There are two good economic reasons to subsidise low carbon generation in the face of competition with fossil fuel-based generation. First, because it is in its relative infancy and hence has not benefitted from the cumulative learning that fossil fuel technologies have enjoyed, subsidies can be justified because of their future learning benefits.87 Second, because fossil fuels produce environmental pollutants, such as particulates, acid rain and carbon dioxide, clean low carbon technologies can justify additional financial support, which reflects the value of reducing these pollutants. A further problem of low carbon generation is that because of the nature of the cash flows associated with such investments, there are high upfront costs and lower future running costs, relative to fossil fuels. This means that long-term power purchase contracts are more valuable for low carbon generation in order to reduce the cost of capital and improve the relative net present value (NPV) of low carbon investments. These reasons suggest that long-term fixed prices for low carbon generation may be needed to help them earn a return at the current stage of their technological development. This is true for both new technologies such as solar PV and onshore and offshore wind and also for established low carbon technologies such as hydro and nuclear. 87

 See Grubb et al. (2008) and Twomey and Neuhoff (2008).

70 

M. G. Pollitt et al.

Just because low carbon generation must be subsidised relative to fossil fuel generation does not mean it cannot be procured via a competitive process. Clearly it is theoretically desirable to minimise the subsidy costs of a given MWh of clean electricity. This can be done by a suitably designed procurement auction, of the type that we describe in the following text. Another problem that must be addressed is the fact that electricity is more valuable at certain times of the day, week and season. This argues against simple fixed-price contracts for MWhs, which do not vary the price paid with the relative value of the power to the system. A way should be found to give more of an incentive for renewable generation to respond to the supply and demand factors that should drive wholesale power prices. This can be done by a contract for difference (CfD) with the government that guarantees a top-up payment based on average wholesale prices, or via a premium FIT where the low carbon generation participates fully in the wholesale market and receives the market price plus a premium set by the government (rather than just a fixed price).

2.5.1.2  General Reform Experience While many countries have had fixed feed-in tariffs per MWh, a number have used competitive procurement methods to support low carbon generation, with some exposure to real-time price volatility. The most common method is to use is tradable green certificates (TGCs).88 This involves requiring suppliers to source a percentage of their electricity from ‘green’ sources by presenting certificates to show that they have done this. Certificates are created when low carbon generators produce a MWh of electricity. This creates a market for TGCs, which trade at a positive price and provide an additional source of revenue for low carbon generators.89 This exposes the generators to the real-time electricity price and if the certificate market is competitive the price of certificates will reflect the lowest cost way of reaching the target percentage. Such schemes exist in many US states (e.g. New York). One issue with the scheme is that when  See Currier (2013).  See Ciarreta et al. (2014).

88 89

2  Lessons for China from International Experience of Power… 

71

the target percentage is too ambitious and there is a shortage of certificates. This leads to a penalty price binding. This has consistently happened in the UK with the result that the value of the certificates rises to the penalty price, which may be overly generous.90 Procurement auctions allowing the acquisition of low carbon electricity have also been used in the UK and the US. These have been very successful in reducing the price paid for low carbon electricity. The UK had an auction in 2015 to supply low carbon generation from renewables which saw the price paid for onshore wind and solar PV fall dramatically relative to the previously published administratively determined prices (of the order of 20%). The auction was for a 15-year CfD.91 Subsequent auctions in 2017 and 2019 saw dramatic falls in price paid for offshore wind.92 The US has made  use of auctions, particularly for the procurement of small-scale renewables. A good example is the successive rounds of the Renewable Auction Mechanism (RAM) in California that has also seen significant reductions in auction prices for projects of 3–20 MW. These auctions are for fixed prices (not CfDs) but do have up to 50 non-payment hours per year, which means that the incumbent distribution company can curtail the generator off the system when it is not in the interests of the system to run the plant.93

2.5.1.3  Chinese Experience94 In China renewables and nuclear power are paid fixed prices per MWh, with the prices being determined at the provincial level in discussions

 See Pollitt (2012).  See DECC (2015). 92  Prices (in 2012 £) in the auctions for offshore wind fell from £114.39/MWh for delivery in 2018/2019 to £57.50/MWh for delivery in 2022/2023 to £41.61/MWh for delivery in 2024/2025. See https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_ data/file/643560/CFD_allocation_round_2_outcome_FINAL.pdf and https://assets.publishing. service.gov.uk/government/uploads/system/uploads/attachment_data/file/832924/Contracts_for_ Difference_CfD_Allocation_Round_3_Results.pdf 93  Anaya and Pollitt (2015). 94  See Liu et al. (2013); Kahrl et al. (2011a, b) and Chen et al. (2010). 90 91

72 

M. G. Pollitt et al.

between the NDRC and local government.95 There is a national renewables target (which includes nuclear) for the percentage of overall generation that is to come from renewables by the end of the 13th Five-Year Plan period and there are targets for the amount of new nuclear power that the system wishes to add. There are provincial-level non-hydro renewable electricity shares for 2020 (see Table 2.6). However, individual provinces may wish to add renewables and nuclear for GDP growth target reasons or because of local preferences for clean energy (usually driven by favourable weather conditions for renewables). Renewables and nuclear power are not subject to either competitive procurement or direct exposure to wholesale prices. Indeed, renewables Table 2.6  Target share of non-hydro renewable energy in total electricity consumption in 2020

Province Beijing Tianjing Hebei Shanxi Inner Mongolia Liaoning Jilin Heilongjiang Shanghai Jiangsu Zhejiang Anhui Fujian Jiangxi Shandong Henan

Non-hydro renewable share targets (actual value in 2018)

Province

Non-hydro renewable share targets (actual value in 2018)

15% (11.7%) 15% (11%) 15% (11.3%) 14.5% (14.5%) 18% (17.3%)

Hubei Hunan Guangdong Guangxi Hainan

10% (7.5%) 13% (10.2%) 4% (3.5%) 5% (4.2%) 5% (5.2%)

10.5% (11.7%) 16.5% (17%) 20.5% (16.2%) 3% (3.3%) 7.5% (7%) 7.5% (5.3%) 11.5% (11%) 6% (4.9%) 8% (8.6%) 10% (9.4%) 10.5% (9.4%)

Chongqing Sichuan Guizhou Yunnan Tibet Shaanxi Gansu Qinghai Ningxia Xinjiang Total

2.5% (2.9%) 3.5% (4.4%) 5% (4.5%) 11.5% (15.6%) 13% (16.9%) 12% (10.6%) 19% (13.4%) 25% (18.5) 20% (22.3%) 13% (14.7%) 9% (9.2%)

Source: NEA website Available at: http://zfxxgk.nea.gov.cn/auto87/201603/t20160303_2205.htm

 The national benchmark prices for onshore wind and solar PV are reasonably generous at 0.60 RMB/kWh and 0.98 RMB/kWh in 2016. 95

2  Lessons for China from International Experience of Power… 

73

are often constrained off the system by grid constraints or the desire to meet the contractual running hour targets of large fossil fuel power plants. As noted earlier, this is because reducing the hours that renewable generation runs reduces system cost, given the subsidy. This suggests that renewable feed-in tariffs do not currently reflect society’s willingness to pay for renewables (but lie above this). Competitive procurement of renewables to set the price paid to renewables would seem to be desirable for two reasons. First, it removes the negotiated price element that seems to result in higher prices than the society is actually willing to pay for renewables. Second, competitive procurement is a firmer contractual commitment, in that improving grid access for renewables will directly reduce the prices in the procurement auction. There is a need for experimentation with competitive procurement as this would be an unusual process for the government to use to achieve its objectives in an industry dominated by state-owned enterprises. There are many companies in the generation sector and there is clearly a lot of opportunity for competitive bidding should the auctions be carefully designed to deliver a competitive outcome. Auctions over wider areas (several provinces) and across different technologies (wind and solar) will highlight the value of location and of different technologies in ways that the current technology and provincially differentiated tariffs do not. Currently none of the market pilot projects involve experimenting with the competitive procurement of renewables. There would seem to be clear opportunity for a pilot project in renewables procurement. The Chinese government announced an intention to introduce a new tradable green certificate scheme form July 2017.96 This has now been introduced.97 According to Chinese official statistics, under collection of renewable levies has led to an accumulated deficit in the renewable payments fund that has, in turn, delayed payments to renewable generators. This payment deficit had grown to 50 billion RMB in January 2017.98  ‘China to launch green certificates for renewable power in July’, Retrieved from http://www. reuters.com/article/us-china-economy-renewables-idUSKBN15I0AK 97  See http://www.nea.gov.cn/2017-02/06/c_136035626.htm 98  Please refer to http://www.jyjch.com/xzlw/2017/0119/4746.html 96

74 

M. G. Pollitt et al.

2.5.2 C  ost Reflective Access Terms for Renewables (12) 2.5.2.1  Theoretical Significance The location of renewables is a particular issue because while fossil fuel plants can be located close to load centres or where transmission capacity is readily available, renewables need to be located where the underlying resources are available. An electricity system with a high penetration of renewables is one where concentrations of loads and generators are often separated by great distances, and where small-scale renewables may be located across the network. Given that the cost of delivering power is a combination of its generation cost and its transportation cost, the pricing of the transportation element faced by generators is very important. Locational signals on where to connect renewables into the electricity grid are complicated by the fact that renewables are currently subsidised. This means that locational signals may—correctly—penalise the connection of renewables in places where generation conditions are very favourable (i.e. remote areas which are windy and/or sunny). This produces the counter-intuitive result that we would be prepared to pay more for a MWh of wind or solar power generated where conditions are less favourable than where conditions are more favourable.

2.5.2.2  General Reform Experience Many international jurisdictions have chosen to pay renewables the same per MWh regardless of location and not to expose them to differential connection charges that reflect the costs to the system of connecting them in particular locations. Indeed, most electricity systems simply socialise the costs of offering firm connection (100% guaranteed export capacity) to renewable generators, so that renewable generators only pay for their direct connection costs (the works to physically connect them to the existing grid or the so-called shallow connection costs). This is now beginning to change as renewables shares increase significantly in some jurisdictions. The RAM auctions in California rank

2  Lessons for China from International Experience of Power… 

75

projects after including the costs to the transmission system of absorbing a given project99 (see Anaya and Pollitt 2015). In the UK renewable generators do pay a share of the upgrade costs to the system of their connection (they have to pay to be connected at an unconstrained part of the network or bear the costs of upgrading the first substation to which they would be connected, the so-called shallowish connection costs). The Flexible Plug and Play project in the UK offered non-firm connection to generators in return for paying shallow connection charges and exposing them to the risk of interruption. For small renewables projects embedded in a distribution system built only to supply loads this resulted in significant total project cost savings100 (Anaya and Pollitt 2015). Signalling where to connect and exposing renewables to the system cost of their connection to constrained parts of the network are in its infancy. Given the intermittent nature of renewables, the high fixed costs of renewables projects (which means that maximising MWhs supplied is important) and the high fixed costs of offering firm grid export capacity to renewables this would seem to be important.

2.5.2.3  Chinese Experience The Chinese power grid is already under significant stress even with relatively low penetration of intermittent (i.e. non-hydro) renewables due to long-distance power flow constraints (see Ming et al. 2016). The potential for long-distance power flows is significantly increasing as the volume of renewables grows. Renewable resources are significantly located in the north and west, a long way from the demand centres in the east and south.101 This suggests significant value in small-scale renewables located close to loads. At the moment generators are not directly exposed to price signals that indicate the value of their connection to the network at particular  See Anaya and Pollitt (2015).  See n. 96. 101  See Hove and Mo (2016). See also Guo (2014). 99

100

76 

M. G. Pollitt et al.

locations. They are indirectly exposed to this via their exposure to curtailment. As this curtailment is arbitrary it is not clear that this properly signals the true value of connection at particular parts of the network.102 Network costs are significant in China and have the potential to rise further. There is, currently, a desire to build out all the constraints in the electricity network and accommodate the projected generation capacity. However, this looks challenging in the face of sustained growth in demand and the increasing use of renewable generation (and indeed fossil generation) located further from loads. Offering efficient connection signals to renewables would seem to be valuable given the scales involved. This suggests that renewable generators should be exposed to some locational signals that indicate the value of their connection at particular locations. This could involve a combination of zonal payments for annual connection capacity to the transmission system for transmission-­ connected generation (following zonal transmission charges in the UK)103 and contributions to distribution system upgrade costs for distribution-­ connected generation that wants higher guaranteed export capacity from the grid.104

2.5.3 A  ppropriate Pricing of Environmental Externalities (Both Carbon Dioxide and Other Atmospheric Pollutants, such as Sulphur Dioxide) (13) 2.5.3.1  Theoretical Significance The production of electricity from fossil fuels is associated with significant environmental externalities. These include the production of nitrous oxides (NOX), sulphur dioxide  (SO2)  and carbon dioxide, as well the production of particulates. Nitrous oxides and sulphur dioxide

 See Zhang et al. (2015b) and Zhang and Li (2012).  See Pollitt and Bialek (2007). 104  See Pollitt and Anaya (2016). 102 103

2  Lessons for China from International Experience of Power… 

77

contribute to acid rain, carbon dioxide to global warming and particulates to dangerous local pollution inter alia. Given the presence of different electricity generation technologies with different environmental characteristics, it is important to price these external costs in such a way as to signal the relative value of clean production to society. Environmental pollution taxes, permit schemes and legal liability for damage105 can all be used to efficiently reflect external costs back on to the producers who create pollution. This raises the costs of fossil fuel-­ based production relative to nuclear and renewables cost.

2.5.3.2  General Reform Experience There have been very positive experiences with combining reformed wholesale markets with appropriate pricing of environmental externalities. This has led to big improvements in the quality of the environment and lower imposed system costs. Three good examples are the US experience with a national sulphur dioxide permit scheme, the South Coast Air Quality Management District (SCAQMD) RECLAIM scheme for the pricing of nitrous oxides and the EU pricing of carbon dioxide. The US sulphur permit scheme started in 1994106 by requiring all large coal-fired power plants in the US to produce permits for each tonne of sulphur dioxide that they produced. This scheme successfully reduced SO2 produced from these plants by 60% by 2000 at very low cost by incentivising the introduction of new cheaper flue gas desulphurisation (FGD) equipment and by switching to low sulphur coal. The RECLAIM scheme was established in 1994107 and is a permit scheme for pricing nitrous oxides in Southern California (covering Los Angeles). This involves pricing nitrous oxides to reflect the atmospheric

 See Viscussi et al. (2005) for a good introduction.  See Ellerman (2003). 107  See Fowlie et  al. (2012) available at: http://nature.berkeley.edu/~fowlie/fowlie_holland_mansur_reclaim.pdf 105 106

78 

M. G. Pollitt et al.

conditions over the area covered. This has been successful in managing the amount of local pollution in the area. The EU introduced a permit scheme for carbon dioxide in 2005. This covers the power sector and a number of other energy-intensive industrial sectors (and was recently extended to aviation). The power sector is around 60% of the current scheme. The scheme has resulted in supporting the switching between coal- and gas-fired electricity generation where this was necessary to stay within the permitted quantities of CO2 in the permit market (see Koenig 2011).

2.5.3.3  Chinese Experience China has a significant problem with air pollution from power plants. In 2014 approximately 30%, 28% and 5% of SO2, NOX and particulate matter (PM) came from the power sector.108 Small-scale coal-fired power plants have been shut down and replaced with much more thermally efficient larger plant.109 There has been a move towards installing FGDs on all new coal-fired power plants. It has thus far made very limited use of pricing to reduce this air pollution to socially optimal levels. However there have been moves to improve the price-based incentives towards cleaner production from fossil fuel power plants.110 The NDRC does allow higher prices to plants with FGDs and seven carbon market pilots have been introduced which each cover the power sector and work in a similar way to the EU ETS.111 The pricing of carbon dioxide in the pilot projects has been $1–2 per tonne of CO2. This is well short of the $18.60 that might be required to facilitate switching from coal- to gas-­ fired power generation in 2015  when considering a new  power plant investment  (see Appendix I), let alone the price required to switch between a current gas and a current coal plant which would be more like $48.61 (see Appendix II).  Source: China Electricity Council and National Bureau of Statistics.  See Wei et al. (2011) and Dupuy et al. (2015). 110  See China National Energy Administration (2012) 111  See Zheng (2016). See also Yu et al. (2014). 108 109

2  Lessons for China from International Experience of Power… 

79

There is some limited pricing of sulphur dioxide in some jurisdictions. Within Shanghai, emissions of SOX and NOX are taxed at 4000 RMB/ tonne.112 Regulated generation prices include a premium for production of electricity from coal in the presence of an FGD unit.113 However, this premium is simply provided for the presence of FGD-equipped power plants without consition of their pollutant emission reduction performance. Therefore, some power plants installed the low-cost, poor-­quality FGDs just to earn the benefits from the price premium. Moreover, one study found that up to 40% of those generation units equipped with FGDs did not use them (see Chow and Perkins 2014). China announced that it intends to introduce a national carbon dioxide permit market in 2017, but it is yet to be implemented. This is a key part of its Intended Nationally Determined Contribution (INDC) submitted to COP-21.114 This will cover the electricity sector and be a significant addition to the global effort on greenhouse gas emissions reduction. This can be gradually tightened to begin to encourage more efficient coal use and switching from coal to gas in the power sector. A key recommendation would be the need for China to start using market-based mechanisms to price local air pollution, whose costs are both time- and location-dependent. A national SO2 market covering both the power sector and other industrial sectors would seem to be a good idea, while the idea of local air resources boards (following the experience of RECLAIM) to price local contributions to smog and particulates would also be very valuable.115

 See Yuan (2016).  The national premium level for production of electricity from coal with FGD is 0.0015 RMB/KWh during the period from 1 July 2007 to 1 May 2014. In the latest policy, the government gives a premium of 0.001 RMB/KWh (connected to network before 1 June 2016) and 0.0005 RMB/KWh (connected to network before 1 June 2016) to coal-fired power plants. Sources: http://www.nea.gov.cn/2014-04/04/c_133235649.htm; http://www.hebwj.gov.cn/News. aspx?sole=20160104163510593; http://www.cec.org.cn/xinwenpingxi/2011-08-25/65025.html; http://www.cec.org.cn/xinwenpingxi/2011-08-25/65025.html; http://www.sdpc.gov.cn/fzgggz/ jggl/zcfg/201404/t20140403_615508.htm 114  Available at http://newsroom.unfccc.int/unfccc-newsroom/china-submits-its-climate-actionplan-ahead-of-2015-paris-agreement/. See also Jackson et al. (2015) and Sha et al. (2015). 115  See Chen et al. (2016). 112 113

80 

M. G. Pollitt et al.

2.6 A  ll Good Power Market Reforms (and Indeed, Significant Economic Reforms More Generally) Involve Appropriate Transition Mechanisms (14) 2.6.1 Theoretical Significance Economic theory traditionally emphasises end-points and equilibrium outcomes in optimal policy design. It suggests that what is, simply, required is to redesign the whole system in a particular way and that this will deliver a set of desirable outcomes. The problem in power market reform is that the system is large, complex and somewhat unpredictable. Much can go wrong in the course of even the best-intentioned and best-­implemented reforms. More importantly given that many reform elements cannot be costlessly implemented immediately, there will— for many years—be a situation where not all of the elements for a theoretically robust power market reform will have been implemented. The system will be ‘out of equilibrium’ and will need to worry about whether such an ‘out of equilibrium’ situation is better than where it started from. Many economists have suggested that the sequencing of reforms is therefore very important116 as well as the transitional arrangements, particularly to protect the key stakeholders that the society is interested in (such as poorer residential customers or small businesses).

2.6.2 General Reform Experience Many electricity reform processes have not gone as society has expected them to. The Californian electricity reforms of 1996 proved disastrous, resulting in rolling blackouts by 2000–01. As we have already observed, this was partially a result of the combination of transitional arrangements (fixed retail prices and a restriction on the use of hedging contracts by incumbent retailers) and opportunistic behaviour by competing generators in the face of high demand.  See Aghion and Blanchard (1994).

116

2  Lessons for China from International Experience of Power… 

81

Even in successful reform countries, such as the UK, there have been problems with market power in wholesale generation market in the 1990s (see Newbery 2005) and a lack of competition in the retail market for inert retail customers since 2008 (see CMA 2016). Many countries, including the UK, have made use of transitional arrangements. These have included offering default regulated tariffs even after the market has opened up to competition. The retail electricity market in Great Britain was fully liberalised from 1999, but incumbent retailers had price controls on their standard tariffs in place until 2002. In Northern Ireland, the retail margin charged by the incumbent electricity retailer was still subject to regulation in 2016, in spite of being technically opened up to competition in 1999. This suggests that transitional arrangements can last for some time.

2.6.3 Chinese Context China is well experienced at transitional arrangements that restrict full competition and protect incumbent firms and existing customers.117 Such arrangements are particularly important in economies where so many prices continue to be regulated and where a ‘big bang’ approach where all prices are simultaneously deregulated would produce significant economic dislocation. Full wholesale market competition would significantly favour low-cost producers of fossil fuels. While this is desirable in the long run it would result in immediate significant losses of output and revenue for certain generation plants and consequent reductions in coal demand from certain mines. Here transitional arrangements might include capacity payments for power plants to keep them open and maintain profitability.118 Significant retail competition would also result in losses for the incumbent grid companies. This would be premised on them having set the levels of their network charges correctly to continue to allow their operations to be financed.  See Mathews and Tan (2013) and Kahrl et al. (2011a, b).  See Menezes and Zheng (2016). See also China National Development and Reform Commission and National Energy Administration (2015) 117 118

82 

M. G. Pollitt et al.

China is still some way from moving to full wholesale and retail competition. Even where market segments (e.g. large industrial users of electricity) are opened up to competition there is a case for some transitional arrangements (such as maximum prices) or restrictions on the ability of generators/retailers to cross-subsidise competition by increasing prices elsewhere. Transitional arrangements can be introduced selectively and where necessary. Good monitoring of the effects of the introduction of prices on the bills consumers pay and the profitability of state-owned companies is necessary in order to assess whether transitional arrangements are adequate or necessary. A key task of the reform process would seem to be reducing the volume of new investment going into the power sector. As we have already noted this was $120 bn in 2015, at a time when demand growth was beginning to mote. Reducing this level of investment will require the phased redeployment of the huge resources being put into to both new power station building and network asset expansion. Even if this amount could be reduced by only $10 bn per year, this would potentially save around 2.5% of the industrial price of electricity.119 China has instituted a significant number of pilot projects (see Table 2.7), but many of these do not test many of design principles of liberalised markets we have outlined in this paper. For example, experiments to test spot market for power in real time, competitive procurement for renewables and locational pricing signals for renewables are missing from the set of pilot projects. Indeed, it is not altogether clear what is really being tested in the pilot projects beyond the systems and ways of working in price-based arrangements, since the principles of liberalisation are well established from global experience. From Table 2.7, we are also interested in the role of local governments’ energy agencies for advancing these pilot projects. The literature points out that local energy agencies in China do not have a wide range of decision-­making powers and policy space to launch their own pilot projects. Most of the current pilot projects are designed by the NDRC (though with participation from the provincial and local energy agencies). The central and local NDRCs supervise the provincial energy agencies to  In all $10 bn out of industrial electricity revenue of $402 bn in 2014.

119

2  Lessons for China from International Experience of Power… 

83

Table 2.7  Electricity market pilot projects in 2016 Selected local pilots in China Shenzhen City

Inner Mongolia

Ningxia Hui Autonomous Region

Yunnan Province

Progress summary National Development and Reform Commission (NDRC) launched Shenzhen City’s Transmission-Distribution Price reform pilot in November 2014. NDRC had a detailed check on the transmission-distribution price, and after implementing this reform pilot, the level of transmission-­ distribution price even decreased, which also lowered the terminal power-sale price. During this reform period, a number of power-selling companies were set up and the commercial electricity price became lower in Shenzhen City. The industrial and commercial utilities both share the benefits of a lower power price. In June 2015, the Transmission-Distribution Price reform pilot was approved in Inner Mongolia by NDRC. NDRC also reviewed the revenue of power transmission and distribution operations as well as the price in Inner Mongolia in September 2015. This review clarifies the transmission-distribution prices of different voltage levels and for different customers. Basically, this reform indicates that customers should pay different prices based on the voltage levels and the price should also include cross-­ subsidisation. As a result, in this reform scheme the reduced price has mainly brought benefits to the large-scale industrial users. NDRC completed its review on Ningxia’s Scheme of Transmission-Distribution Price reform pilot in September 2015. Ningxia project is also the first pilot reform approved by the State Grid. Compared to other electricity reform schemes in Inner Mongolia and Shenzhen City, the Ningxia Scheme made a clear progress in setting up mechanisms for transmission-distribution price reform. Yunnan Province is a pioneer in China’s market-oriented electricity trading. In 2015, the Industry & Information Technology Commission in Yunnan and Yunnan provincial government established a ‘3134’ trading mode, which covers ‘three main parts, one center, three markets, and four modes’. Based on these market-oriented trading and transmission-distribution reform progresses, NDRC reviewed the Scheme of Transmission-Distribution Price reform pilot of Power Grid in Yunnan Province in October 2015, and Yunnan was approved as a comprehensive power trading pilot in November 2015. (continued)

84 

M. G. Pollitt et al.

Table 2.7 (continued) Selected local pilots in China Guizhou Province

Shanxi Province

Chongqing

Guangdong Province

Progress summary In July 2015, the pilot scheme—Deepening Power Sector Reform—in Guizhou province was approved by Guizhou government. It adopts a similar model as Shenzhen City for reforming the transmission-distribution price. There are four drivers for this reform pilot in Guizhou: (i) thermal power companies face difficulties because of the reduced utilisation hours of power generation; (ii) Guizhou provinces has abundant coal resources; (iii) high electricity consumption—local industries need lower price electricity supply and (iv) Guizhou province had strong foundation for direct electricity trading. The Shanxi pilot was approved by Reform Commission NDRC and NEA in February 2016. Three features of this pilot are: (i) the Shanxi Grid independently runs a trading centre and connects with other shareholders together; (ii) this pilot reform sets up a clear incremental distribution standard and (iii) it has established a spot trading mechanism.

The pilot in Chongqing was approved by the General Office of NDRC and NEA’s General Affairs Department in November 2015. In December 2015, three pilot selling companies were set up, which marked the beginning of Chongqing’s pilot. In February 2016, 12 companies entered an agreement with one selling electricity corporation, which shows that the authority has put the pilot in practice and currently keeps refining it. Similar to Chongqing, the selling side reform in Guangdong is however in a slower progress. In 2016, the authority established the basic rules of the trading mechanism and enhanced the technical supporting system by simulation operations. The electricity trading was planned to officially start in 2017.

Source: China5e Research Centre (2016, pp. 34–61)

implement these local pilots. More freedom to initiate local pilot projects, for instance within the China Southern Grid provinces, with support from local governments and industry stakeholders would seem desirable.

2  Lessons for China from International Experience of Power… 

85

Xu (2017, p.  170) emphasises the point that previous trials with market competition prior to March 2015 were not successful because of a lack of generation capacity in the trial, below-cost retail prices, insufficient interconnection and a lack of experienced regulatory oversight. This highlights the importance of timing and adequate preparation for a sustained and successful reform process.

2.7 Conclusions 2.7.1 International Lessons and Policy Priorities for China Successful reform in China is not about the success of its electricity companies as companies (consider the US and Germany), except in the sense that really efficient electricity companies support the rest of the economy by releasing labour to be more productive elsewhere and keeping the cost of power down. The entire utility privatisation programme in the UK between 1979 and 1997 released 2% of the entire workforce back to the rest of the economy and boosting economy-wide productivity. A key driver of the current reform is the high price of electricity for industrial customers relative to the US. We have identified four major savings within the power sector that would bring down prices for industrial customers. These are reform of dispatch (which might reduce coal use by up to 6% and allow industrial prices to fall by 1–2%), increasing the efficiency of the grid companies (which might reduce industrial prices by 2–3%), rebalancing charges away from industrial to residential customers to better reflect underlying system costs (which might reduce industrial prices by up to 5%) and reducing the high rate of investment in generation/networks by $10 bn per year which could also reduce prices for industrial customers by the order of 3%. None of the savings are easy to deliver because they have significant redistributional implications. However, they have been achieved in many other countries, albeit over a period of up to ten  years. They suggest that the non-fuel cost gap— amounting to 12% of the current Chinese industrial electricity price— that we identified between China and the US can be eliminated.

86 

M. G. Pollitt et al.

If the gap with the US is to be further reduced, this would take a comprehensive reform of the coal sector (and of value added taxation in the electricity sector). Rationalisation of the coal sector might reduce costs to those in the US, and this would substantially close the remaining price gap. A combination of tax changes or cheaper sources of energy (e.g. shale gas) could further reduce the price differential. China needs to view electricity market reform in the context of what it can do for the rest of the Chinese economy and resist vested interests within the sector that would seek to limit its rationalisation. A key part of this is the opportunity to simultaneously rationalise the coal production sector (which has 4.3 m employees, slightly more than the whole electricity sector), by reducing coal demand and improving coal sector productivity. The decoupling of electricity policy from national procurement strategies for fossil fuel and nuclear technologies was a key driver of cost reduction towards new investment in Europe and the US. An additional impetus to reduce Chinese dependence on coal for power production might be the rapid recent decline in the reserves to production ratio for Chinese coal. Comprehensive power market reform based on the creation of competitive wholesale and retail markets and separately regulated network businesses is once again being pushed forward in China (following the publication in March 2015 of the No.9 document), building on the 2002 reform (which separated the grid from generation but subsequently stalled). The primary objective is to lower prices for industrial customers, with the additional objectives of reducing wind curtailment and reducing overinvestment in new, dirty coal-fired power plants. The likely extent of power market reform remains linked to reduction in coal use. Unless there is a willingness to rationalise and reduce coal use in China at the individual plant/mine level, the power market reform will make limited progress on the ground. China has devolved a lot of energy investment decisions to the provinces. This has favoured provincial coal mines and encouraged the pursuit of energy independence among the provinces. This is because coal production and coal generation contribute to provincial GDP targets, and local coal mines and coal generation contribute to provincial tax revenue. This undermines a regional/national market emerging to the extent that

2  Lessons for China from International Experience of Power… 

87

is fully beneficial to the national economy. Clearly the central government has to strongly regulate interprovincial electricity trade and encourage its development. While much of the reform process emphasises a move to market trading of electricity, power plants are still not being dispatched in merit order in real time. The government needs to prioritise reform of dispatch (with its implications for coal use and the value of existing generation assets) if a genuine wholesale market is to emerge which drives operational costs down and incentivises significant efficiency gains from making better use of the existing power generation fleet. The government’s current capacity to regulate a competitive power sector is limited. There is a shortage of well-qualified/well-trained staff (accountants, economists and lawyers), who can administer and regulate the institutions of the market, partly due to low public sector salaries (relative to SOE salaries). There continues to be a need to reduce the power of the State Grid Corporation in setting/frustrating policy in favour of well-resourced and independent (of the industry) civil servants. This could be achieved by transferring some of the research functions of State Grid to the central government, and treating State Grid as an interested party with its own internal financial incentives (like generators) in policy discussions. More encouragingly, the latest round of reforms (embodied in the No.9 document) has laudable intentions but also is proceeding rather carefully given the complicated and interconnected nature of unwinding the current regulatory arrangements surrounding the power sector. A number of pilot projects introducing wholesale forward markets for industrial power are underway and the government is, rightly, proceeding cautiously towards comprehensive market reform. There has been real progress in separating out generation/retail from the network businesses and provincial prices for network access (based on assessments of the regulatory asset and operating cost base) are now published. Many hundreds of electricity retail companies have been created, but none are competing directly with SGCC and CSG (as we discuss in the next chapter). The preconditions for incentive regulation of the network business of State Grid and Southern China Grid are, however, now in place.

88 

M. G. Pollitt et al.

The period from 2015 has (so far) been a good time to push forward with reform. Final prices started high (for industry and this is a major driver of reform because of low US industrial energy prices). The electricity industry in 2015 was profitable relative to underlying costs (which have fallen in line with commodity price falls). The pilot wholesale markets are showing price reductions (as we discuss further in the next chapter) for industrial customers. There is also environmental pressure to end wind curtailment (which remains high) and mostly due to the allocated hour-based contracts held by coal-fired power plants. The moment may of course pass if commodity prices rise consistently.

2.7.2 Suggestions for Future Research We have identified five sources of lower power prices for industrial users. These are all worthy of several multifaceted studies. 1. The size of the dispatch savings. This requires careful modelling of constraints and of the distributional implications for individual companies and consition of what compensation is necessary; 2. Modelling of the impacts of reallocating charges from industrial to residential customers over a ten-year period, during which incomes and household consumption are expected to continue to grow strongly; 3. Efficiency modelling of the scope for cost reduction in the grid companies and exposure to incentive regulation. There is a surprising lack of papers on the efficiency of the different business units of the grid companies; 4. Modelling of the financial impact on electricity customers of overinvestment in generation and consition of how new investment in fossil fuel generation (to the extent that it is required) can be reduced; 5. Further work on the scope for rationalising the fuel input sector in China and how China can get access to cheaper coal and gas. This might draw parallels from historical heavy industry rationalisation processes in Europe, Korea and Japan.

2  Lessons for China from International Experience of Power… 

89

In addition, we would note the following: There needs to be investigation of the design and lessons from the market pilot projects. The current pilot projects need to demonstrate that they are supporting the underlying rationalisation of the sector that is required to bring its costs down and improve its environmental performance. There might be a need for additional pilots to trial different aspects of reform (such as price bid-led dispatch). The calculation of network tariffs needs to be carefully analysed. This benefits from independent study because the regulator is initially in a weak information position relative to the regulated companies. The scope for network tariffs to be calculated wrongly  or to be miss-­ calibrated is quite high. China has taken a number of significant reform steps since 1985. It is, perhaps, surprising that there has not been more careful analysis of the impact of the different reform steps. For instance, how have publicly owned generators performed against privately owned ones, or whether different state investment vehicles have a better record of managing their assets efficiently. In particular, it would also be good to look at the impact of the 2002 reorganisation of generation and distribution on efficiency more closely. China needs to benchmark its performance with those of other countries. This is particularly valuable in the initial stages of reform, when it is difficult to get meaningful internal benchmarks. It would be good to see more comparative analysis of the costs of generation and networks between other countries and China. China’s power market reform still needs to address its air pollution problem. While rationalisation of the power sector will reduce pollution, a low carbon transition is still required and this will necessitate close attention to how renewable and nuclear procurement can be made more competitive. There should be more attention to comparative research between sectors within China: it would be good to consider whether successful liberalisation and economic regulation of other sectors might inform electricity reform in China. Regulation and liberalisation are transferable skills between sectors and there may be things to learn from, for instance, the IT sector in China.

90 

M. G. Pollitt et al.

There needs to be a new institutional economic analysis of how to design a successful regulator in China (drawing on lessons from other countries and sectors) by considering the incentives within the civil service to effectively regulate large monopoly companies in a way that ensures a reasonable degree of independence from arbitrary central and provincial government interference. The question of how best to make use of the ability for different regions to move at different speeds should be carefully considered (e.g. reforming residential pricing in richer provinces first). Research that is specific to the reform circumstances of particular provinces would be valuable. Finally, given the large degree of misunderstanding (globally) of the macro benefits of lower power prices and the release of labour from the power and coal system, it would be good to see some more general equilibrium modelling of the benefits to the Chinese economy of power market reform.

Appendices  ppendix I: Calculation Process of the Switching A Carbon Prices from Coal- to Gas-Fired Power Plant Investment in 2015 (LCOE=levelised cost of electricity) No. Steps

Coal

Gas

Source

1

77.72

92.79

IEA (2015, p. 98)

0.07772

0.09279

1

1

2

LCOE (discount rate 7%)(USD/ MWh) LCOE (discount rate 7%)(USD/ KWh) LCOE difference (USD/KWh) (Coal-gas) Fuel consumption

−0.01507(0.07772 –0.09279) 442.216(g/KWh)

0.19125 (Nm3/KWh) Zhang et al. (2012, p. 232) (continued)

2  Lessons for China from International Experience of Power… 

91

Appendix 1  continued No. Steps

Coal

2

Emission coefficient (coal)

2.78124 kg CO2/kg Coal

2

Emission coefficient (gas)

2.19362 kg CO2/Nm3

2

CO2 emission (kg CO2/KWh) CO2 emission (tonne CO2/ KWh) CO2 emission difference (kgCO2/KWh) (coal-gas) Switch Price from coal to gas (USD/ tonne)

1.230 (442.216∗2.78124/1000) 0.00123

2

2

3

Gas

Source Zhang et al. (2012, p. 233) Zhang et al. (2012, p. 233)

0.420 (0.19125∗2.19362) 0.000420

0.000810 (0.001230–0.000420)

18.60 (0.01507/0.000810)

 ppendix II: Calculation Process of the Switching A Carbon Prices from Coal- to Gas-Fired Power Generation in 2015 Industrial electricity price (US $/KWh) In 2014 US 0.0710 China 0.1068

Coal price (US $/KWh) in 2014

Gas price (US $/KWh) in 2014

Switching price $/tonne CO2

0.0241 0.0384

0.0159 0.0778

6.98 48.61

92 

M. G. Pollitt et al.

China Coal

Gas

(g/KWh) 442.216 yuan/tonne

(Nm3/KWh) 0.19125 yuan/m3

534

2.5

Carbon emissions

kg CO2/kg Coal 2.78124

kg CO2/Nm3 2.19362

Power generation cost difference (yuan/KWh) Carbon emissions difference (tonne/KWh) Switching CO2 price ($/tonne)

−0.241981656

Heat rate Price

0.000810379 48.61

US Heat rate Price Carbon emissions Power generation cost difference ($/KWh) Carbon emissions difference (tonne/KWh) Switching CO2 price ($/tonne)

Coal

Gas

(g/KWh) 442.216 $/tonne 54.5 kg CO2/kg Coal 2.78124

(Nm3/KWh) 0.19125 $/m3 0.1555995 kg CO2/Nm3 2.19362

−0.005657632 0.000810379 6.98

References English ACER (Agency for the Cooperation of Energy Regulators). (2015). ACER Annual Market Monitoring Report 2015. Retrieved from http://www.acer. europa.eu/Official_documents/Acts_of_the_Agency/Publication/ACER_ Market_Monitoring_Report_2015.pdf

2  Lessons for China from International Experience of Power… 

93

Aghion, P., & Blanchard, O. J. (1994). On the speed of transition in Central Europe. NBER Macroeconomics Annual, 9, 283–330. Alva, H. A. C., & Li, X. (2018). Power sector reform in China: An international perspective. Paris: IEA Publications. An, B., Lin, W., Zhou, A., & Zhou, W. (2015). China’s market-oriented reforms in the energy and environmental sectors. Paper presented at the Pacific energy summit. Retrieved from http://nbr.org/downloads/pdfs/ETA/ PES_2015_workingpaper_AnBo_et_al.pdf Anaya, K. L. (2010). The restructuring and privatisation of the Peruvian electricity distribution market. Cambridge: Faculty of Economics, University of Cambridge. Anaya, K. L., & Pollitt, M. G. (2015). Options for allocating and releasing distribution system capacity: Deciding between interruptible connections and firm DG connections. Applied Energy, 144, 96–105. Archer, C. L., Simão, H. P., Kempton, W., Powell, W. B., & Dvorak, M. J. (2017). The challenge of integrating offshore wind power in the US electric grid. Part I: Wind forecast error. Renewable Energy, 103, 346–360. Bergman, L., Brunekreeft, G., Doyle, C., von der Fehr, N.-H. M., Newbery, D. M., Pollitt, M., & Regibeau, P. (1998). A European Market for Electricity? London: Centre for Economic Policy Research. Bessant-Jones, J.  E. (2006). Reforming power markets in developing countries: What have we learned? Energy and Mining Sector Board Discussion Paper No. 19, September 2006. Washington, DC: World Bank. Bishop, M., Kay, J., & Mayer, C. (1994). Privatization and economic performance. Oxford: Oxford University Press. Bohn, R. E., Caramanis, M. C., & Schweppe, F. C. (1984). Optimal Pricing in Electrical Networks Over Space and Time. RAND Journal of Economics, 18(3), 360–376. BP. (2019). Statistical Review of World Energy 2019. London: BP. CEC (China Electricity Council). (2013). Electricity industry statistics summary 2012. Internal report. Chao, H. P., & Peck, S. (1996). A market mechanism for electric power transmission. Journal of Regulatory Economics, 10(1), 25–59. Chawla, M., & Pollitt, M. (2013). Global trends in electricity transmission system operation: Where does the future lie? The Electricity Journal, 26(5), 65–71. Chen, Q., Kang, C., Xia, Q., & Zhong, J. (2010). Power generation expansion planning model towards low-carbon economy and its application in China. IEEE Transactions on Power Systems, 25(2), 1117–1125.

94 

M. G. Pollitt et al.

Chen, H., Tang, B. J., Liao, H., & Wei, Y.-M. (2016). A multi-period power generation planning model incorporating the non-carbon external costs: A case study of China. Applied Energy, 183, 1333–1345. Chen, H., Chyong, C. K., Mi, Z. F., & Wei, Y. M. (2019). Reforming the operation mechanism of Chinese electricity system: Benefits, challenges and possible solutions. The Energy Journal, 41(2), 191–217. Cheng, C., Chen, F., Li, G., Ristic, B., Mirchi, A., Qiyu, T., & Madani, K. (2018). Reform and Renewables in China: The architecture of Yunnan’s hydropower dominated electricity market. Renewable and Sustainable Energy Reviews, 94, 662–693. Chow, G. C., & Perkins, D. H. (2014). Routledge handbook of the Chinese economy. Oxford: Routledge. Ciarreta, A., Espinosa, M. P., & Pizarro-Irizar, C. (2014). Switching from feed­in tariffs to a tradable green certificate market. In The interrelationship between financial and energy markets (pp. 261–280). Berlin Heidelberg: Springer. CMA. (2016). Energy Market Investigation Final Report. London: Competition and Markets Authority. CNESA (China Energy Storage Alliance) (2015), China’s New Electric System Reforms. Retrieved from http://en.cnesa.org/featured-stories/2015/8/4/ chinas-new-electric-system-reforms Crossley, D. (2014). Energy efficiency as a resource for the power sector in China. Beijing: Regulatory Assistance Project. Cubbin, J., & Stern, J. (2006). The impact of regulatory governance and privatization on electricity industry generation capacity in developing countries. The World Bank Economic Review, 20(1), 115–141. Cunningham, E. A. (2015). The state and the firm. Retrieved from: http://ash. harvard.edu/files/chinas-energy-working-paper.pdf Currier, K. M. (2013). A regulatory adjustment process for the determination of the optimal percentage requirement in an electricity market with tradable green certificates. Energy Policy, 62, 1053–1057. Danish Energy Regulatory Authority. (2014). 2014 National Report to the European Commission. http://www.ceer.eu/portal/page/portal/EER_HOME/ EER_PUBLICATIONS/NATIONAL_REPORTS/National%20 Reporting%202014/NR_En/C14_NR_Denmark-EN.pdf DECC (Department for Business, Energy & Industrial Strategy). (2015). Contracts for difference (CFD) allocation round one outcome. Retrieved from https://www.gov.uk/government/statistics/contracts-for-difference-cfdallocation-round-one-outcome

2  Lessons for China from International Experience of Power… 

95

Domah, P., & Pollitt, M. G. (2001). The restructuring and privatisation of the electricity distribution and supply businesses in England and Wales: A social cost-benefit analysis. Fiscal Studies, 22(1), 107–146. Dupuy, M., Weston, F., & Hove, A. (2015). Stronger markets, cleaner air, power sector – Deepening reform to reduce emissions, improve air quality and promote economic growth. Retrieved from: http://www.paulsoninstitute.org/wp-content/uploads/2015/09/2-Power-Sector-EN-Final.pdf Ellerman, D. (2003), Ex post evaluation of tradable permits: The U. S. SO2 cap-­ and-­trade program. MIT CEEPR Working Paper 2003–03. Retrieved from: http://ideas.repec.org/p/mee/wpaper/0303.html Ellerman, A. D., Marcantonini, C., & Zaklan, A. (2016). The European Union emissions trading system: Ten years and counting. Review of Environmental Economics and Policy, 10(1), 89–107. Florio, M. (2004). The great divestiture: Evaluating the welfare impact of the British privatizations, 1979–1997. Cambridge, MA: MIT press. Fowlie, M., Holland, S. P., & Mansur, E. T. (2012). What do emissions markets deliver and to whom? Evidence from Southern California’s NOx trading program. The American Economic Review, 102(2), 965–993. Gilbert, R., & Newbery, D. M. (1994). The dynamic efficiency of regulatory constitutions. RAND Journal of Economics, 25(4), 538–554. Grubb, M., Jamasb, T., & Pollitt, M. G. (2008). Delivering a low-carbon electricity system  – Technologies, economics and policy. Cambridge: Cambridge University Press. He, Y.  X., Yang, L.  F., He, H.  Y., Luo, T., & Wang, Y.  J. (2011). Electricity demand price elasticity in China based on computable general equilibrium model analysis. Energy, 36, 1115–1123. Hogan, W. (1992). Contract networks for electric power transmission. Journal of Regulatory Economics, 4(3), 211–242. Hove, A., & Mo, K. (2016). Going for gold: Championing renewable integration in Jing-Jin-Ji. Retrieved from http://www.paulsoninstitute.org/wp-content/ uploads/2016/07/Renewable-Energy-Integration-EN.pdf International Energy Agency [IEA]. (2015). Projected costs of generating electricity: 2015 edition. Paris: OECD. International Energy Agency [IEA]. (2016). CO2 emissions from fuel combustion: 2016 edition. OECD: Paris. Jackson, R. B., Canadell, J. G., Le Quéré, C., Andrew, R. M., Korsbakken, J. I., Peters, G. P., & Nakicenovic, N. (2015). Reaching peak emissions. Nature Climate Change, 6, 7–10.

96 

M. G. Pollitt et al.

Jamasb, T., & Pollitt, M. (2005). Electricity market reform in the European Union: Review of progress toward liberalization & integration. The Energy Journal, 29 (Special Issue on European Electricity Liberalisation), 11–42. Jamasb, T., & Pollitt, M. (2007). Incentive regulation of electricity distribution networks: Lessons of experience from Britain. Energy Policy, 35(12), 6163–6187. Joskow, P. L. (2003). The difficult transition to competitive electricity markets in the US. Mimeo. Retrieved from: http://economics.mit.edu/files/1160 Joskow, P. L. (2008). Lessons learned from electricity market liberalization. The Energy Journal, Vol.29 Special issue No.2, The future of electricity: Papers in honor of David Newbery, 9–42. Joskow, P. L., & Schmalensee, R. (1983). Markets for power: An analysis of electric utility deregulation. Cambridge, MA: MIT Press. Kahrl, F., & Wang, X. (2014). Integrating renewables into power systems in China: A technical primer – Power system operations. Beijing: Regulatory Assistance Project. Available at: http://www.raponline.org/document/download/id/7459 Kahrl, F., Williams, J., & Ding, J. (2011a). Four things you should know about China’s electricity system. Retrieved from https://www.wilsoncenter.org/publication/four-things-you-should-know-about-chinas-electricity-system Kahrl, F., Williams, J., Jianhua, D., & Junfeng, H. (2011b). Challenges to China’s transition to a low carbon electricity system. Energy Policy, 39(7), 4032–4041. Kahrl, F., Williams, J. H., & Hu, J. (2013). The political economy of electricity dispatch reform in China. Energy Policy, 53, 361–369. Kahrl, F. F., Dupuy, M., & Xuan, W. (2016). Issues in China power sector reform: Generator dispatch. Retrieved from http://www.raponline.org/knowledgecenter/issues-in-china-power-sector-reform-generator-dispatch/ Khalid, O. (2016). Exploring the market for demand-side response. 13 May 2016 Spring Seminar. Available ­at: http://www.eprg.group.cam.ac.uk/wp-content/ uploads/2016/05/O.Khalid.pdf Kim, T.-H. (2016). The Central Electricity Generating Board (CEGB) and the politics of fuels and power generation 1961–89. PhD Thesis, University of Cambridge. Koenig, P. (2011). Modelling Correlation in Carbon and Energy Markets, EPRG Working Paper No.1107. Lei, N., Chen, L., Sun, C., & Tao, Y. (2018). Electricity market creation in China: Policy options from political economics perspective. Sustainability, 10(5), 1–15.

2  Lessons for China from International Experience of Power… 

97

Li, Y., Lukszo, Z., & Weijnen, M. (2016). The impact of inter-regional transmission grid expansion on China’s power sector decarbonization. Applied Energy, 183, 853–873. Littlechild S. (2006). Foreword: The market versus regulation. In F. P. Sioshansi & W. Pfaffenburger (Eds.), Electricity market reform: An international perspective. Oxford: Elsevier. Littlechild, S. C. (2008). Municipal aggregation and retail competition in the Ohio energy sector. Journal of Regulatory Economics, 34(2), 164–194. Liu, Z., Guan, D., Crawford-Brown, D., Zhang, Q., He, K., & Liu, J. (2013). Energy policy: A low-carbon road map for China. Nature, 500(7461), 143–145. Liu, Y., Eyre, N., Darby, S., Keay, M., Robinson, D., & Li, X. (2015). Assessment of demand response market potential and benefits in Shanghai. Available at: http://www.nrdc.cn/phpcms/userfiles/download/201509/01/Assessment of Demand Response Market Potential and Benefits in Shanghai.pdf Liu, S., Yang, Q., Cai, H., Yan, M., Zhang, M., Wu, D., & Xie, M. (2019). Market reform of Yunnan electricity in southwestern China: Practice, challenges and implications. Renewable and Sustainable Energy Reviews, 113, 1–24. Ma, J. (2011). On-grid electricity tariffs in China: Development, reform and prospects. Energy Policy, 39(5), 2633–2645. Mansur, E. T., & White, M. W. (2012). Market organization and efficiency in electricity markets. Retrieved from: http://www.dartmouth.edu/~mansur/ papers/mansur_white_pjmaep.pdf Mathews, J. A., & Tan, H. (2013). The transformation of the electric power sector in China. Energy Policy, 52, 170–180. Menezes, F. M., & Zheng, X. (2016). Regulatory incentives for a low-carbon electricity sector in China. Retrieved from http://www.uq.edu.au/economics/ abstract/562.pdf Ming, Z., Ximei, L., & Lilin, P. (2014). The ancillary services in China: An overview and key issues. Renewable and Sustainable Energy Reviews, 36, 83–90. Ming, Z., Lilin, P., Qiannan, F., et al. (2016). Trans-regional electricity transmission in China: Status, issues and strategies. Renewable and Sustainable Energy Reviews, 66, 572–583. Mingtao, Y. A. O., Zhaoguang, H. U., & Zhang, N. (2015). Low-carbon benefits analysis of energy-intensive industrial demand response resources for ancillary services. Journal of Modern Power Systems and Clean Energy, 3(1), 131–138.

98 

M. G. Pollitt et al.

Mota, R. L. (2003). The restructuring and privatisation of electricity distribution and supply businesses in Brazil: A social cost-benefit analysis. Cambridge: Department of Applied Economics, University of Cambridge. Newbery, D. (2002). Privatization, restructuring, and regulation of network utilities (Vol. 2). Cambridge, MA: MIT Press. Newbery, D. (2005). Electricity liberalisation in Britain: The quest for a satisfactory wholesale market design. The Energy Journal, 26(Special Issue: European Electricity Liberalisation), 43–70. Newbery, D. (2016). Towards a green energy economy? The EU Energy Union’s transition to a low-carbon zero subsidy electricity system–Lessons from the UK’s Electricity Market Reform. Applied Energy, 179(October), 1321–1330. Newbery, D., Strbac, G., & Viehoff, I. (2016). The benefits of integrating European electricity markets. Energy Policy, 94, 253–263. Nillesen, P. H., & Pollitt, M. G. (2011). Ownership unbundling in electricity distribution: Empirical evidence from New Zealand. Review of Industrial Organization, 38(1), 61–93. O’Donnell, A.  J. (2003). The Soul of the Grid: A Cultural Biography of the California Independent System Operator. Lincoln: iUniverse Inc. Ofgem. (2009). Regulating Energy Networks for the Future: RPI-X@20 Performance of the Energy Networks under RPI-X. London: Ofgem. Pingkuo, L., & Zhongfu, T. (2016). How to develop distributed generation in China: In the context of the reformation of electric power system. Renewable and Sustainable Energy Reviews, 66, 10–26. Pollitt, M.  G. (2005). Electricity reform in Chile: Lessons for developing countries. Retrieved from https://www.researchgate.net/publication/227578069_ Electricity_Reform_in_Chile_Lessons_for_Developing_Countries Pollitt, M. G. (2008a). Electricity reform in Argentina: Lessons for developing countries. Energy Economics, 30, 1536–1567. Pollitt, M. G. (2008b). The future of electricity (and gas) regulation in a low-­ carbon policy world. The Energy Journal, 29(Special Issue), 63–94. Pollitt, M. G. (2009). Electricity Liberalisation in the European Union: A Progress Report. Mercato Concorrenza Regole, 3/2009, 497–523. [In Italian]; English Version: EPRG Working Paper, No. 0929. Pollitt, M. G. (2012). Lessons from the history of independent system operators in the energy sector. Energy Policy, 47, 32–48. Pollitt, M. G., & Anaya, K. L. (2016). Can current electricity markets cope with high shares of renewables? A comparison of approaches in Germany, the UK and the State of New York. The Energy Journal, 37(Special Issue), 69–88.

2  Lessons for China from International Experience of Power… 

99

Pollitt, M. G., & Bialek, J. (2007). Electricity network investment and regulation for a low carbon future. Retrieved from https://www.repository.cam.ac.uk/ handle/1810/194734 Pollitt, M., & Stern, J. (2011). Electricity regulation in developing countries: Developments since 2001. Utilities Policy, 19, 53–60. Rahimi, A. F., & Sheffrin, A. Y. (2003). Effective market monitoring in deregulated electricity markets. IEEE Transactions on Power Systems, 18(2), 486–493. Rioux, B., Galkin, P., Murphy, F., & Pierru, A. (2016). How do price caps in China’s electricity sector impact the economics of coal, power and wind? Potential gains from reforms. Working Paper KS-1652, King Abdullah Petroleum Studies and Research Center (KAPSARC), P.O. Box 88550, Riyadh 11672, Saudi Arabia, Retrieved from https://www.kapsarc.org/wp-content/ uploads/2016/09/KS-1652-DP047-Potential-Gains-from-Reforming-PriceCaps-in-Chinas-Power-Sector.pdf Sha, F., Ji, Z., & Linwei, L. (2015). An analysis of China’s INDC. Beijing: China National Center for Climate Change Strategy and International Cooperation. Sheffrin, A. (2002, December). Predicting market power using the residual supply index. In FERC Market Monitoring Workshop, Retrieved from www.caiso. com/docs/2002/12/05/2002120508555221628.pdf Shleifer, A. (1985). A theory of yardstick competition. The RAND Journal of Economics, 16(3), 319–327. Sioshansi, S., Fereidoon, P., & Wolfgang, P. (2006). Electricity market reform: An international perspective. Amsterdam: Elsevier. Slaughter and May. (2016). Competition law in China  – Slaughter and May. Available at: https://www.slaughterandmay.com/media/879862/competition-law-in-china.pdf Stoft, S. (2002). Power system economics Designing Markets for electricity. Piscataway: IEEE press. Sweeney, J. L. (2002). California Electricity Crisis. Stanford, CA: Hoover Press. Taylor, N. W., Jones, P. H., & Kipp, M. J. (2014). Targeting utility customers to improve energy savings from conservation and efficiency programs. Applied Energy, 115, 25–36. Twomey, P., & Neuhoff, K. (2008). Will the market choose the right technologies? In Delivering a low-carbon electricity system – Technologies, economics and policy. Cambridge: Cambridge University Press. Viscusi, W. K., Harrington, J. E., & Vernon, J. M. (2005). Economics of regulation and antitrust. Cambridge, MA/London: MIT Press.

100 

M. G. Pollitt et al.

Wang, Q., & Chen, X. (2012). China’s electricity market-oriented reform: From an absolute to a relative monopoly. Energy Policy, 51, 143–148. Wang, J., Bloyd, C. N., Hu, Z., & Tan, Z. (2010). Demand response in China. Energy, 35(4), 1592–1597. Wei, Y., Liu, L., Wu, G., & Zou, L. (Eds.). (2011). Energy economics: CO2 emissions in China. Berlin/London: Springer. Wei, Y., Chen, H., Chyong, C. K., et al. (2018). Economic dispatch savings in the coal-fired power sector: An empirical study of China. Energy Economics, 74, 330–342. Weiss, L.  W. (1975). Antitrust in the electric power industry. In A.  Phillips (Ed.), Promoting competition in regulated markets. Washington, DC: Brookings Institution. Xu, Y. C. (2017). Sinews of power: Politics of the State Grid Corporation of China. Corby: Oxford University Press. Yao, M., Hu, Z., Sifuentes, F., et al. (2015). Integrated power management of conventional units and industrial loads in China’s ancillary services scheduling. Energies, 8(5), 3955–3977. Yeo, Y. (2008). Regulatory politics in China’s telecommunications service industry: When socialist market economy meets independent regulator model. Paper presented at the (Re)Regulation in the Wake of Neoliberalism: Consequences of Three Decades of Privatization and Market Liberalization. Yu, S., Wei, Y.-M., Guo, H., & Ding, L. (2014). Carbon emission coefficient measurement of the coal-to-power energy chain in China. Applied Energy, 114, 290–300. Zhang, C., & Heller, T.  C. (2007). Reform of the Chinese electricity power market: Economics and institutions. In D. Victor & T. C. Heller (Eds.), The political economy of power sector reform: The experiences of five major developing countries. Cambridge: Cambridge University Press. Zhang, S., & Li, X. (2012). Large scale wind power integration in China: Analysis from a policy perspective. Renewable and Sustainable Energy Reviews, 16(2), 1110–1115. Zhang, D., Liu, P., Ma, L., et al. (2012). A multi-period modelling and optimization approach to the planning of China’s power sector with consition of carbon dioxide mitigation. Computers & Chemical Engineering, 37, 227–247. Zhang, O., Yu, S., & Liu, P. (2015a). Development mode for renewable energy power in China: Electricity pool and distributed generation units. Renewable and Sustainable Energy Reviews, 44, 657–668. Zhang, S., Jiao, Y., & Chen, W. (2017). Demand-side management (DSM) in the context of China’s on-going power sector reform. Energy Policy, 100, 1–8.

2  Lessons for China from International Experience of Power… 

101

Zheng, D., & Zhou, W. (2003). A design for regional ancillary services auction markets in China. Paper presented at the IEEE Bologna PowerTech Conference, Bologna. Zheng, W., J. Hu, M. Zhang and M. Miao, Improved autonomous converter control for multi-terminal HVDC networks considering line impedance effect in long-distance interconnections, 12th IET International Conference on AC and DC Power Transmission (ACDC 2016), Beijing, 2016, pp. 1–6. Zhou, H., Su, Y., Chen, Y., Ma, Q., & Mo, W. (2016). The China southern power grid: Solutions to operation risks and planning challenges. IEEE Power and Energy Magazine, 14(4), 72–78.

Chinese China Electricity Council. (2015). Annual electricity statistis 2015. Available at: http://www.cec.org.cn/ China National Development and Reform Commission (NDRC). (2010). “关 于印发《电力需求侧管理办法》的通知.” Demand Side Management Measures, Document No. 2643, November 4, 2010, Retrieved from: http:// www.ndrc.gov.cnfzgggz/jjyx/dzxqcgl/201011/t20101116_381342.html China National Development and Reform Commission and National Energy Administration (2015, March 23). “发展改革委、能源局关于改善电力 运行调节促进清洁能源多发满发的指导意见.” NDRC, NEA: Improving power operations, adjusting incentives for clean energy production guiding opinion. Retrieved from: http://www.gov.cn/xinwen/2015-03/23/content_2837637.html China National Energy Administration. (2012, January 4). “电力发展规划、 应当体现合理利用能源、电源与电网配套发展、提高经济效益和有 利于环境保护的原则.” Power Development Planning: Principles for Appropriately Using Energy, Power and the Grid System; Raising Economic Efficiency; and Benefiting Environmental Protection. Retrieved from: http:// www.nea.gov.cn/2012-01/04/c_131262818.htm China National Energy Administration. (2015, December). “能源局就推进电 力市场建设的实施意见答记者问. Retrieved from: http://www.nea.gov. cn/2015-11/30/c_134869326.html China National Energy Administration (NEA) (2016). Regulatory report of transmission and distribution cost of grid companies in six provinces (冀北 等6省(地区)电网企业输配电成本监管报告). Available at: http://zfxxgk. nea.gov.cn/auto92/201606/t20160614_2264.htm

102 

M. G. Pollitt et al.

China Southern Power Grid (CSPG). (2015). Corporate social responsibility report 2015. Available at: http://www.csg.cn/ China State Council. (2015, March 21). “关于进一步深化电力体制改革的 若干意见.” Deepening reform of the power sector, Document No. 9. Retrieved from http://www.ne21.com/news/show-64821.html China5e Research Centre. (2016). “中国电改试点进展政策研究与建议” The Policy Study of China’s Power Sector Reform Pilot Project Development and Recommendations. China5e Research Centre Report. Beijing. Feng, Y.-C. (2016a). “从云南方案看新电改隐患” Yunnan local pilot and Challenges for China’s recent power sector reform. China Energy. Retrieved from http://www.cnenergy.org/yw/zc/201602/t20160205_270260.html Feng, Y.-C. (2016b). “如何理解中国电力体制改革:市场化与制度背景” How to understand China’s power sector reform: Market reforms and institutional context. Retrieved from http://www.chinareform.org.cn/economy/ price/refer/201611/t20161130_25845 Guo, Q. (2014, February 25). “国家能源局召开风电产业监测沟通会.” China National Renewable Energy Centre, Retrieved from http://www. cnrec.org.cn/hd/2014-02-25-412.html National Bureau of Statistics (NBS). (2015). China energy statistical yearbook 2015. Beijing: China Statistics Press. National Bureau of Statistics (NBS). (various years). Statistical yearbook of China. Beijing: Chinese Statistics Press. State Grid Corporation of China (SGCC). (2015). Corporate social responsibility report 2015. Retrieved from: http://www.sgcc.com.cn/ Tan, R.  Y., & Zhao, G.  H. (2016). 中国能源监管探索与实践. In China’s energy governance: Foundation and practice. Beijing: People’s Publishing House. Yuan C.H. (2016), "燃煤发电项目的经济性研究.  中国国情国力"  China National Conditions and Strength, 10: 61-65. Retrieved from: https://www. greenpeace.org.cn/wp-content/uploads/2016/04/中国煤电的经济性.pdf Zhang, X. et al. (2015b). “售电侧市场放开国际经验及其启示” International experience and lessons in power sales side market liberalization. State Grid Energy Research Institute. Retrieved from http://www.aeps-info.com/ aeps/ch/reader/create_pdf.aspx?file_no=20151128001&flag=1&journal_ id=aeps&year_id=2016 Zheng, X. (2016). 突破"不可能三角" 中國能源革命的緣起、目標與實現 路徑. In China’s energy Trilemma: China’s energy revolution, goals and reform paths. Beijing: Beijing Sciences Press.

3 Power Market Reform in China: Lessons from Guangdong In Collaboration with Chung-Han Yang and Hao Chen

3.1 Introduction Guangdong is the largest and the most economically successful province in China. In 2017 Guangdong contributed more than 27.1% of Chinese exports, 10.9% of GDP, 8.1% of the population (c.112 m) and 9.5% of electricity consumption in China.1 Guangdong has relatively high final electricity prices in China (for residential and most industrial and commercial customers) and is a net importer of power from other provinces.2 It contains the Shenzhen special economic zone, which allows the introduction of new market measures not currently rolled out across China. Guangdong has been leading the way within China on power market reform.3 The province is part of the region of the China Southern Grid (CSG), which has been a dynamic and innovative area of the national electricity system since its creation in 2002.4 Even before the No.9  These data are drawn from the website of National Bureau of Statistics(NBS), http://www. stats.gov.cn/ 2  See Cheng (2016). 3  See IEA (2019, p. 38). 4  Chau et al. (2011). See also Wen (2017). 1

© The Author(s) 2020 M. G. Pollitt, Reforming the Chinese Electricity Supply Sector, https://doi.org/10.1007/978-3-030-39462-2_3

103

104 

M. G. Pollitt et al.

document was published an electricity market pilot project had begun in Shenzhen in 2014.5 This consisted of the publication of separate electricity transmission and distribution charges and the introduction of monthly contract trading of electricity between certain generators and retail customers. In 2016, two power exchange centres were established. The Guangdong Power Exchange6 looks after the power market in Guangdong, while the Guangzhou Power Exchange facilitates interprovincial trading of electricity across the CSG area.7 This chapter seeks to document and analyse progress with the introduction of wholesale and retail power markets in Guangdong in the light of international experience. We will build our discussion on the previous chapter. Well-functioning power markets are at the heart of delivering successful power market reform, as has been recognised by Stoft (2002). Appropriate competition between generators and retailers should in theory lead to the realisation of the key recommendations that we made in the last chapter. Thus, our study of the reforms in Guangdong will focus on the following questions. What are the key achievements of recent power market reform in Guangdong to date? How has market piloting changed the traditional payment and power station dispatch systems? How have transmission and distribution charges been calculated and how are they being regulated? To what extent is the market pilot impacting on the current cross-subsidies within the electricity system? How is the market pilot impacting on operational and investment decisions both within the generators and the network companies? What progress is being made in integrating interconnector flows into the electricity market? What progress is being made on creating a full set of electricity markets?

 China National Development and Reform Commission (2015a). ‘国家发展改革委关于深圳市 开展输配电价改革试点’ 发改价格【2014】2379号Retrieved from: http://jgs.ndrc.gov.cn/ zcfg/201411/t20141104_639639.html 6  See https://pm.gd.csg.cn 7   Guangzhou Power Exchange Center (2017a). 《南方区域跨区跨省月度电力交易规则 (试行), Retrieved from: https://www.gzpec.cn/main/index.do 5

3  Power Market Reform in China: Lessons from Guangdong 

105

This chapter aims to assess progress with reform, and what Guangdong is learning about how electricity market models need to be adapted for its own particular circumstances. A complete set of electricity markets can be easily stated, but in practice jurisdictions across the world have developed their own sets of electricity markets (e.g. PJM in the US is different to the market in Great Britain). The chapter highlights what the lessons from the market pilot experiences in Guangdong are for both the province itself and for the rest of China. The chapter draws on the experience of Chinese stakeholders, to identify what are the key problems to be overcome in bringing about a successful electricity reform transition in the World’s most significant electricity system. The chapter offers some recommendations for next steps in the reform process at the provincial level and is intended to be a positive contribution to ongoing debates about the detailed implementation of electricity sector reform in China and to be a platform for future discussion and informed input on the appropriateness of international reform experience in the Chinese context. The rest of the chapter is organised as follows. In Sect. 3.2, we begin with a discussion of the background to the reforms in Guangdong, including a discussion of the characteristics of the power system. Section 3.3 discusses how the power market pilot actually works in Guangdong and whether the current market design is in line with power markets we see elsewhere. Section 3.4 explores the extent to which power market reform has brought new players into the electricity system in Guangdong. Section 3.5 considers the effects of the reform on the operational and dispatch decisions of firms in the sector. In Sects. 3.3, 3.4, and 3.5 we aim to provide some international context (expanding on Chap. 2) as background to our analysis of the reform effects observed to date. Section 3.6 offers some points for improvement in the light of the existing market design and its observed effects.

106 

M. G. Pollitt et al.

3.2 Background 3.2.1 Guangdong Within China Guangdong (see Fig. 3.1 for a map) plays an important role in the development of policy for the whole of China.8 It has been a leader in pro-­ market institutional developments, such as a clearer system of law and governance. Guangdong’s capital is Guangzhou, one of the largest cities in the world by population (at around 15 million, and its GDP ranks fourth in all the cities in China), while its second city, Shenzhen

Fig. 3.1  Map of Guangdong. (Source: https://wikitravel.org/upload/shared/7/7a/ Guangdong2.png)

 Bui et al. (2002) and also Andrews-Speed (2013).

8

3  Power Market Reform in China: Lessons from Guangdong 

107

(population around 12 million, and its GDP ranks third in all the cities in China. Both Guangzhou and Shenzhen are among the top five largest cities in China). Furthermore, Shenzhen is opposite Hong Kong and a Special Economic Zone within China, which has grown from a market town of 30,000 in 1979.9 Guangdong is located around the Pearl River Delta (or Greater Bay Area, which includes Guangzhou, Shenzhen, Macau and Hong Kong) through which much of the exports of the region flow. Shenzhen sits outside the remit of the provincial government of Guangdong and has its own regulatory institutions. Good examples of leadership in governance include Shenzhen’s leading role in the development of financial regulation and Guangzhou’s hosting of one of China’s three intellectual property courts.10 Guangdong is politically significant for the whole of China, because it acts as an economic laboratory with respect to market reforms for the whole of China (in a somewhat similar way to California in the US). Three of five city special economic zones in China—Shenzhen, Zhuhai and Shantou—are located in Guangdong. In addition, it is the neighbour of two special administrative regions of Hong Kong and Macao. China introduced a carbon emissions pilot in 2012.11 Two provinces and five cities were selected as pilots, Guangdong (located in Guangzhou) and Shenzhen being two of them. The seven governments involved could decide on what sectors were included in the pilot. Only Guangdong has some auctioning of its allocation of permits. The Guangzhou Emissions Exchange (GZX) trades three products: GDEAs, China CERs (Certified Emission Reductions) and provincial CERs.12 Ten per cent of allowances can come from provincial CERs and 30% of CERs can come from other provinces. Electricity, cement, petrochemicals and steel sectors were initially covered, with aviation and paper-making added in 2016.13 A  See also Xinhua Finance (2015). Shenzhen given nod to pilot new power transmission, Available at: http://en.xfafinance.com/html/Industries/Utilities/2015/40163.shtml 10  See Reinhold Cohen (2015). 11  Cheng et al. (2016). 12  Cheng et al. (2016). See also Liu, D. (2017). ‘电力市场、碳排放权市场和绿色证书市场的 协调发展’.Electricity market, carbon market and green certificate mechanisms development, industry perspective, China Electrical Equipment Industry, 2017.07. 13  See ICAP Status Report 2017. 9

108 

M. G. Pollitt et al.

national carbon market had been planned for December 2017 covering both electricity and heat sectors.14 The price fell from 60 RMB in 2013 to 12 RMB in 2017, exhibiting similar trend to the EU ETS. There is an annual 20,000 tonnes threshold for participation. The price of carbon trading permit is expected to start around 30 RMB and increase to 200 RMB by 2030. There is an expectation of 3–5 years of overlap between pilots and national market, with no thought of linking. Lessons learned in Guangdong’s carbon market pilot could be taken up in the design of the national market (see Wang et  al. 2016).15 There is a big overlap between players in carbon allowances, electricity and renewables certificates markets. Analysis by Cheng et al. (2016) shows that a significant carbon price (of around $10 per tonne of CO2) would reduce coal use for power generation in Guangdong significantly (but increase the use of natural gas). There are also likely to be significant air pollution co-­benefits (see Cheng et al. 2016). Guangdong is just one of the pilot power markets in China.16 Other significant market pilots exist with differing degrees of coverage and discounts. Electricity prices are high in Guangdong due to lack of cheap gas and longer distances for coal transportation, because of which carbon prices will increase electricity prices more than elsewhere.17 A timeline of significant recent power sector reform steps, relevant to Guangdong, is shown in Fig. 3.2. The retail market in Guangdong is being opened up gradually, with the largest customers by voltage level being offered the chance to buy their power in the power market.18 The process of registration for retailers and generators is the responsibility of the Economics and Information Commission (EIC) (see later in the chapter). A monthly wholesale power  See Pike and Zhe (2017).  Wang et al. (2016). 16  See China5e Research Centre (2016); China National Development and Reform Commission (2015b). ‘电力体制改革解读’. Analysis of Electricity Institutional Reform, Remin Publisher, Beijing. 17  Zhang (2017). See also Zhang and Xu (2017). ‘中国能源大省电力市场建设经验分析’. Electricity Market Construction Experience of Major Provinces of Energy in China. Electric Power, 2017, 50(4): 7–10. 18  See People’s Government of Guangdong Province (2017a). ‘广东电力市场交易基本规则 (试行)’ Retrieved from: http://www.nea.gov.cn/2017-01/20/c_135999956.htm 14 15

3  Power Market Reform in China: Lessons from Guangdong 

109

Fig. 3.2  Timelines for reform in the Guangdong electricity sector. (Source: Adapted from An Bo et al. (2015, p. 6). Revised from Guangzhou Power Exchange (2017b))

market officially started trading in June 2016. As of August 2017, 310 retailers and 60 generators were registered in the market, of which 101 retailers were participating.19

3.2.2 The Size of the Electricity Sector in Guangdong The electricity sector in Guangdong is large, as shown in Fig. 3.3. Total production in 2016 was 404 TWh and capacity installed was 105 GW (both larger than the UK in 2016). Total demand in 2017 was 595 TWh and Guangdong was a significant importer of electricity from other provinces (particularly Yunnan). Electricity demand in Guangdong has been growing rapidly (see Fig. 3.4).20 Demand grew by 7.2% p.a. between 2006 and 2014. Demand growth slowed to 1.5% in 2015, before averaging 6.0% p.a. from 2016 to 2018. The rate of construction of new power generation capacity was 10.15 GW in 2015 and 5.4 GW in 2016. The grid also continues to  See Guangzhou Power Exchange Center (2017b).  See also Yang et al. (2017). ‘广东电力市场需求侧响应交易机制研究’. Research on Demand Response Trading Mechanism in Guangdong Electricity Market, Guangdong Electric Power, 30(5), 25–34. 19 20

110 

M. G. Pollitt et al.

Capacity type in 2016 (GW) 4.17, 3.92%

0.3, 0.28%

3.25, 3.05% 14.57, 13.68%

58.11, 54.56%

10.47, 9.83%

15.63, 14.68%

Coal

Gas

Nuclear

Hydro

Wind

Solar and others

Biomass

Fig. 3.3  Generation capacity in the Guangdong electricity sector. (Source: Guangzhou Power Exchange Centre 2017a)

expand rapidly. A total of 7274 km of new network length (of 220 kV and above) was added in 2015 and 4542 km of new lines were added in 2016.21 The quality of service has been improving from a low base (see Fig. 3.5), especially in the urban centres. Table 3.1 compares Guangdong’s electricity prices to those of Texas at the outset of the reform. Texas has the largest state electricity demand in the US (398 TWh in 2016) and the highest state industrial electricity demand (more than twice that of second-place (in terms of total demand) California). Table 3.1 shows the major driver of reform: the high price of industrial electricity relative to international competitors, such as the US. The final industrial price in Guangdong is significantly higher than in Texas. Some of this differential is to do with the price of natural gas (the marginal fuel in Texas) for power generation in the US versus the price of coal (the marginal fuel in Guangdong) for power generation in  Across the whole of the China Southern Grid area grid investment amounted to 77.5 bn RMB in 2016. 21

3  Power Market Reform in China: Lessons from Guangdong 

111

Electricity consumption (TWh) 600 550 500 450 400 350 300 250 200 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 Fig. 3.4  Electricity demand in Guangdong. (Source: (1) Data of the period from 2006 to 2016 are drawn from National Bureau of Statistics of China, http://www. stats.gov.cn/. (2) Data of 2017 are drawn from Guangdong power exchange centre: https://pm.gd.csg.cn)

Guangdong (around 25% of the 8.4 US cents/kWh). Most of the gap however is not explained by cost differences between Texas and Guangdong. By contrast, the price of residential electricity in Guangdong is lower than the price of industrial electricity and is cheaper than the residential electricity in Texas. Electricity demand (see Fig. 3.6) in Guangdong is predominantly from industry (65%) with only a minority from residential consumers (16%). This is in sharp contrast to the developed countries, for example in 2016 in Texas 37% of demand is from residential consumers and only 28% from industry.22 22

 Source: US EIA.

112 

M. G. Pollitt et al.

Total Hours lost per customer

45 40

City consumers

35

Rural consumers

30 25 20 15 10 5 0 2011

2012

2013

2014

2015

2016

2017

Fig. 3.5  Quality of service in Guangdong. (Source: Compilation of power industry statistics of China (2011–2017))

Table 3.1  Electricity price and fuel input price differential with US

Texas Guangdong Guangdong minus Texas

Industrial electricity price (US $/ kWh) in 2015

Residential electricity price Gas price for Coal price for generation (US generation (US (US $/kWh) in $/kWh) in 2015 $/kWh) in 2015 2015

0.0554 0.1394 0.0840 (152% higher)

0.0161 0.0311 0.0150

0.0094 0.0884 0.0790

0.1167 0.1084 −0.0083 (7% lower)

Sources: Fridley, David, Hongyou Lu, and Liu Xu. Key China Energy Statistics 2016. Berkeley, CA: Lawrence Berkeley National Laboratory, 2017 (p. 28); Guangdong NDRC website: http://www.sz.gov.cn/szzt2010/zdlyzl/sfxx/bz/jg/index.htm; Zhang et al. (2014); EIA; OECD energy price & taxes (pp. 317–18); 1 USD = 6.2284 RMB

3  Power Market Reform in China: Lessons from Guangdong 

113

Electricity consumption in 2018 (TWh) 11.26, 2% Agriculture, Forestry, Animal Husbandry and Fishery

99.46, 16%

Industry

120.95, 19% 400.066, 63%

Tertiary industry

Residential

Fig. 3.6 Sources Exchange Center)

of

electricity

demand.

(Source:

Guangdong

Power

3.3 How the Power Market Works 3.3.1 International Context Power markets have evolved gradually in many leading jurisdictions, such as the US and the UK. They have their origin in two fundamental ideas: ‘merit order dispatch’ and ‘power pools’. Within monopoly generators such as EdF in France or the CEGB in England and Wales, power plants were dispatched in (merit) order of their marginal operating cost in order to meet system demand at any point in time.23 System marginal cost was the marginal cost of an additional MWh given the demand on the system at any time and represented the cost of the least expensive plant needed to supply the last MWh to meet demand. In the US, local integrated monopoly power utilities began to trade electricity across their territorial monopoly borders in order to mutually benefit from system savings arising from differences in their system marginal costs, with systems with lower marginal costs able to export power to those with higher marginal  See Chick (2007, pp. 57–83) who discusses the history of marginal cost pricing in the electricity industries of Britain, France and the US. 23

114 

M. G. Pollitt et al.

costs for mutual benefit. The trading platforms to allow this sort of trading were power pools, which eventually became the independent system operators we see today (such as PJM, MISO or ERCOT in the US).24 Power pools of this type were short-run markets which guided plant operation over the hour or the half hour. The power market reforms of the 1990s saw a much deeper development of power markets with the break-up of the ownership of generation plants between multiple owners and the rise of new entrant generators. Wholesale power markets could now be used not just to trade power between systems but from all individual power plants. Power markets did not just cover short-run (day-­ ahead) markets but also contract markets for longer periods. Power markets have been extended from just energy to ancillary services, such as frequency regulation and capacity. Stoft (2002) discusses what a full set of power markets looks like.25 In the UK, for instance, we observe bilateral energy contract markets (for monthly, annual and other periods) and short-term energy balancing markets (down to one hour ahead of real time). We also see markets for ancillary services (e.g. for frequency and short-term operating reserve). A capacity market, for longer-term reserve capacity, has recently been introduced. These power markets are linked in the sense that changes in the supply and demand balance in one market have implications for the pricing in other power markets. Around the time of the earlier 2002 power market reform, which created China Southern Grid and the Big five generators, Andrews-Speed et al. (2003) suggested a mandatory power pool for Guangdong,26 followed by a move to a regional power market. Zeng et al. (2004) discuss what a complete set of electricity markets looks like in the context of Guangdong.27 Lin et al. (2019) discuss how economic dispatch of power plants in Guangdong, as envisaged by the current round of reforms, could reduce production costs by 13% (though they

 See, for example, Hurlbut et al. (2017) who discuss lessons from MISO’s experience for China.  See Stoft (2002, pp. 202–314) who discusses the complete set of energy markets and ancillary services markets. 26  Andrews-Speed et al. (2003). 27  Zeng et al. (2004). See also Bessembinder and Lemmon (2002). 24 25

3  Power Market Reform in China: Lessons from Guangdong 

115

partly do this by increasing CO2 emissions, by causing gas to coal switching). Power markets can be operated by the system operator or they can be operated by separate entities. In the US power markets tend to be operated by the independent system operator (which does not own any generation, retail or network assets). In Europe, a lot of wholesale power is traded across independent power exchanges, which are financial trading platforms, with limited ownership links to other parts of the electricity system (some still have transmission system operators as shareholders). Power exchanges have merged across Europe and are increasingly co-­ optimising their pricing algorithms to improve the efficiency of trading across a wide area. Thus, currently seven regional power exchanges coordinate their day-ahead pricing algorithms via a single trading platform (EUPHEMIA), the so-called market-coupling.28 This can give rise to a single day-ahead price for wholesale energy across around 85% of the European Union’s electricity in a given hour in the absence of any cross-­ border transmission constraints. Efficient power market prices are about the extent to which prices reflect the underlying fundamentals of supply and demand over both the short and the long term and are a good guide to both short-term operation and long-term investment. There are some excellent analyses of how the process of market extension has brought benefits in the short run. For example, ACER/CEER (2019) shows how the process of market-­coupling has increased the percentage of the time in which power flows in the right direction (from low- to high-price areas),29 and Mansur and White (2012) show how market area extension by PJM has similarly improved pricing efficiency at its former borders.30 It is very important to say that in a power market every generator and every retailer that bids up to the market clearing price should be paid that price (in what should be a uniform price auction).31 This is because they  See APX et al. (2013) for a description of EUPHEMIA.  Ibid. 30  Mansur and White (2012). 31  See Stoft (2002, pp. 93–106) for a discussion of auction designs for electricity markets. He argues against using untested market designs. Stoft (2002, p. 101) has a discussion of the limited circumstances under which pay-as-bid might be preferred to a single price auction in electricity. For a 28 29

116 

M. G. Pollitt et al.

are equally valuable in matching supply and demand. Any other outcome would lead to incentives to game the bidding and would reduce the efficiency of the wholesale price determination process as a way of determining which supplies and demands should be matched in the market.

3.3.2 The Power Market in Guangdong The wholesale power market is currently divided into two parts. The first is an annual market where the price is decided once per year and this covers most of the traded electricity. This is composed of a bilateral contract market between the generators and the customers and a smaller annual contract auction market. In 2019, the annual traded quantity was 200 TWh, of which around 60 TWh is in the monthly market. This means that only 9% of total electricity demand (30% of 30%) is in the monthly market. There is a 30% limit on the market share of retailers in the monthly market, though no limits on retailer market shares in the annual market. While the generators must sell their power in the ratio 70:30 between the annual and monthly power markets, retailers are not restricted in how they procure power from the market. The monthly market price in the power market in Guangdong has actually been a market-determined discount on the regulated retail price.32 The maximum discount is −500 basis points (1 point = 0.001Y per kWh). The market covered 4000 large users in 2017. A typical large user might be a large telecoms company or a metals factory. The annual market discount in 2017 for the bilateral contract market was −0.0645 RMB/kWh, with the monthly discount fluctuating around this (see Fig.  3.7). Prices already vary by time of day (peak-average-valley) by +/−0.3 RMB/kWh. In 2018 the market covered 180 TWh. Generators are paid a regulated price for their power by China Southern Grid (CSG) more detailed discussion of discriminatory price auctions versus single price auctions, see Krishna (2010, pp. 173–184). 32  Li and Shen (2016). Guangdong Electric Power Trading Center yesterday officially launched mainly acts as a ‘matchmaker’, Top News. See also South China Energy Regulatory Bureau of National Energy Administration of PRC China (2015). ‘广东省电力大用户与发电企业直接交 易扩大试点工作方案’ Retrieved from: http://nfj.nea.gov.cn/adminContent/initViewContent. do?pk=zwgk1492

117

3  Power Market Reform in China: Lessons from Guangdong 

Nov-19

Sep-19

Jul-19

May-19

Mar-19

Jan-19

Nov-18

Sep-18

Jul-18

May-18

Mar-18

Jan-18

Nov-17

Sep-17

Jul-17

May-17

Mar-17

-0.04

Jan-17

0 -0.02 -0.06 -0.08 -0.1 -0.12 -0.14 -0.16 -0.18 -0.2 Annual bilateral market discount Annual auction contract market discount Monthly auction discount

Fig. 3.7  Power market prices (RMB/kWh). (Source: Guangdong Power Exchange Center, https://pm.gd.csg.cn)

for all the power that they generate and supply to the grid. The power market determines a discount that coal and gas generators are willing to accept on their regulated generation price for the part of their power that is in the market. China Southern Grid uses the market prices in the power market to discount both the price it charges to retail customers in the power market and the price it pays to generators in the power market; it also pays the agreed margins to retailers in the power market.33 These payments are reflected in the following month’s payments/bills. This avoids the need to separate out the transmission and distribution charges that CSG is charging. Separating out the distribution and transmission charges for all customers is quite difficult because of the implicit cross-subsidies between different customer groups within the current retail charges of CSG.  A particular issue in Guangdong is that Pearl River Delta region subsidises economic development in other parts of Guangdong, through paying  Regulatory Assistance Project (RAP) (2016). See also Sung (2017). ‘广东电力市场改革和售电 策略探讨’. Guangdong electricity market reform and sale strategy, Mechanical and Electrical Information, 2017 June, 89–90. 33

118 

M. G. Pollitt et al.

higher electricity charges. As of August 2017, Guangdong had not published its transmission and distribution charges, which need to be approved by the National Development and Reform Commission (NDRC),34 though these were subsequently published. However Shenzhen published its network charges earlier in the reform process because it has its own transmission and distribution grid which made it easier to calculate its underlying cost.35 All meters are currently owned by CSG.  This reduces a substantial potential source of competition and benefit for customers. Customer/ retailer ownership of meters would give incentives for better use of meter data and meter equipment. However it would also necessitate regulation of meter quality and connection to prevent fraud. CSG is still dominant in the electricity supply industry in Guangdong, distributing 970.3 TWh in 2018 and having 537 bn RMB (c. $71 bn)36 of revenue. CSG is 110th in the Fortune 500. It is managing all of the payment risk in the sector because it collects all the revenue and distributes it to the market participants. This is because of an initial worry about whether new retailers could provide the billing infrastructure and financial stability that would otherwise be necessary. Another issue in Guangdong is unregistered generation.37 This is self-­ generation which is being used to bypass the high grid and other charges that are levied on electricity from the grid. This generation is mainly from smaller, dirty coal-fired power plants.  On 7 November 2017, China NDRC just published Guangdong’s transmission and distribution charges for 2017–2019《国家发展改革委关于广东电网2017–2019年输配电价的通知》(发 改价格〔2017〕969号) http://shupeidian.bjx.com.cn/news/20171107/859803.shtml See also China National Development and Reform Commission (2016a). ‘国家发展改革委 国家能源局 关于印发《电力中长期交易基本规则(暂行)》’ 发改能源〔2016〕2784号Retrieved from: http://www.ndrc.gov.cn/zcfb/zcfbtz/201701/W020170112319053238252.pdf See also China National Development and Reform Commission (2017). ‘国家发展改革委关于印发《省级电 网输配电价定价办法(试行)》’ 发改价格〔2016〕2711号Retrieved from: http://www.ndrc. gov.cn/fzgggz/jggl/zcfg/201701/t20170104_834333.html 35  See China National Development and Reform Commission (2015a). ‘国家发展和改革委关于 深圳市开展输配电价改革试点’ 发改价格【2014】2379号Retrieved from: http://jgs.ndrc. gov.cn/zcfg/201411/t20141104_639639.html 36  At $1 = 6.64 RMB. 37  See China Electric Council (2017) 告别无序竞争 电力市场呼吁售电专业化时代http://www. cec.org.cn/xinwenpingxi/2017-10-30/174433.html 34

3  Power Market Reform in China: Lessons from Guangdong 

119

The ultimate goal for power market reform is to open up all industrial and commercial users to competition. It might also be that residential customers could participate in the power market on a voluntary basis. Published grid charges are being recalculated every three  years in line with NDRC pricing formula which includes a fixed price and an inflation adjustment. Spot market operational trials began in May of 2019.38 The spot market design is based on PJM with 15-minute day-ahead bidding and nodal pricing.39 We note that a spot market has been planned in China since at least 2008.40 There is an intention to introduce a frequency regulation market, but not a market for demand-side response.

3.4 New Players 3.4.1 International Context A striking impact of power market reform in many leading jurisdictions is the proliferation of companies actively involved in the trading of electricity (in the widest sense). The creation of wholesale markets is premised on the creation of a separate supply and demand side in the wholesale market. Generators are on the supply side and retailers are on the demand side. For wholesale markets to function efficiently there  China National Development and Reform Commission (2017). ‘国家发展改革委关于印发《 省级电网输配电价定价办法(试行)》’ 发改价格〔2016〕2711号Retrieved from: http:// www.ndrc.gov.cn/fzgggz/jggl/zcfg/201701/t20170104_834333.html China National Development and Reform Commission (2017). 国家发展和改革委 《关于 深化电力现货市场建设试点工作的意见》的通知, 发改办能源规〔2019〕828号, retried from: http://www.ndrc.gov.cn/zcfb/gfxwj/201908/t20190807_943963.html 39  Tung and Huang (2017). ‘美国 PJM 电力市场及对广东电力改革的启示’ Experiences in PJM Market in the United States: A Good Reference for the Power Market Reform in Guangdong Power Grid, Yunnan Electric Power, 2017, 45(1). See also Zhang et  al. (2015), ‘售电侧市场 放开国际经验及其启示’ International Experience and Lessons in Power Sales Side Market Liberalization, State Grid Energy Research Institute. Retrieved from: http://www.aeps-info.com/ aeps/ch/reader/create_pdf.aspx?file_no=20151128001&flag=1&journal_id=aeps&year_id=2016 40   Please see the latest progress of spot market pilots in China at http://www.nea.gov. cn/2017-­09/05/c_136585412.htm See also Tsai et al. (2017). “英国电改专家怎么看待中国电 改 ,给 了 什 么 建 议 ?——专 访 剑 桥 大 学 能 源 政 策 研 究 所 副 所 长 迈 克 尔 ·波 利 特 教 授 ” Interview with Professor Michael Pollitt, Energy Observer, Retrieved from: http://www.eothinker. com/eo/show.php?itemid=569 38

120 

M. G. Pollitt et al.

needs to be multiple generators and multiple retailers. It is not enough for there to be one large generator selling to one large retailer. Thus, in the US and Europe we have seen generation and retail opened up to competition. Large industrial and commercial customers have been allowed to shop around for a retailer/supplier (or indeed are free to set up a retail/trading company directly). Similarly, generators have been able to enter the retail market, that is, to directly acquire final customers for their electricity. Some of the retail companies have been set up by third parties new to the electricity industry, often with experience in gas, telecoms or financial markets. The most successful of these third-party entrants have been those from the gas industry (e.g. British Gas in the UK is the largest new entrant into the electricity sector, with half of all the customers switched to British Gas, or GdF-Suez [the former French gas incumbent] who have been very successful in entering the industrial and commercial electricity markets in Northern Europe). In the UK (which we discuss in more detail in Chap. 4) there has been a proliferation of generators and retailers who participate in wholesale power markets. Immediately prior to the power market reform in 1990 there were 14 retail area monopolies in Great Britain (i.e. one retailer per customer). There were two regional generation companies. The 1990 reform created a market with 6 generators (the CEGB was effectively separated into 4 parts) and 14 former area monopoly retailers. The former gas retail monopoly for Great Britain (British Gas) immediately entered the retail market, as did the four former CEGB generators seeking to acquire final customers directly. New generators could also enter both generation and retail, as could stand-alone retailers. In 2017 there were 149 licensed generators and 68 industrial and commercial retailers in the UK wholesale power market41 and market had very low generator and retailer concentration. Generators and retailers are exposed to significant market risks that need to be managed. Generators have both fixed and variable costs that need to be covered by their sales revenue. Likewise retailers mostly sell power at fixed prices for a year to their customers, with a guarantee to  See Ofgem (2017).

41

3  Power Market Reform in China: Lessons from Guangdong 

121

supply all of their demand. This exposes both parties to significant potential financial risks. A sharp rise in wholesale power prices could bankrupt retailers, while a sharp fall in wholesale prices could bankrupt generators. This encourages both parties to hedge their positions with longer-term fixed-price contracts (for 1–2  years) for most of their generation/sales. This limits their financial exposure to short-term wholesale prices. This happens regardless of what percentage of power is actually traded in short-term markets. In PJM in the US generators have to trade all of their power in a compulsory day-ahead market, while in Great Britain generators trade only around 5% of their power in the near real-time balancing market. In Great Britain cases the degree of bilateral contracting (direct contracts between buyers and sellers) is of the order of 90% of all traded contracts.42 Retailers and generators participate directly in power markets in Europe and North America. Retailers also have to manage their own billing and collection systems. They have to pay relevant government taxes and charges on power, network charges and wholesale costs. Non-­ payment or missing billing is a serious issue, because retail margins (i.e. the difference between all their external costs and the revenue they receive) are small, of the order 5–10% of total revenue. Retail companies have been bankrupted by poor data management and billing (e.g. Independent Energy in the UK in 200043). Generators exposed to low wholesale prices have also been bankrupted (e.g. British Energy44 and TXU Europe45 in 2002 in the UK). Another set of new players in deregulated electricity markets are energy service companies who seek to manage the energy costs of electricity customers. These companies can have a range of business models including owning electricity assets and selling power at a fixed price (rather like a  See Ofgem (2016) which reports that most power that is traded in Great Britain is traded in OTC (over-the-counter) contracts, not via power exchanges. 43  See M. Harrison, ‘Independent Energy collapses with customers still owing £119 m in bills’, The Independent, 8 September 2000. Available at: http://www.independent.co.uk/news/business/news/ independent-energy-collapses-with-customers-still-owing-pound119m-in-bills-699827.html 44  See Taylor (2007). 45  See M.  Harrison, ‘Generators exposed after collapse of TXU Europe’, The Independent, 20 November 2000. Available at: http://www.independent.co.uk/news/business/news/generators-­ exposed-­after-collapse-of-txu-europe-128536.html 42

122 

M. G. Pollitt et al.

conventional generator) or seeking to manage electricity costs through better metering and finding the best market price.46 Energy service companies are often IT-based enterprises that focus on aggregating demands across their customers and seeking the best price for their customer base and then sharing cost savings with them. They would mostly not be exposed to full market price risk in the same way as a conventional retailer, but essentially receive a fee for service. Power market reform is not just about wholesale markets, but it is also about the introduction of incentive regulation of power networks.47 Pressure to cut costs across the electricity industry—in both generation and networks—creates pressure to competitively outsource supply contracts for the creation and operation of new generation and network assets.48 This creates or widens procurement markets. Existing companies often divest their construction businesses and seek to run tenders for new business. For example, many of the distribution companies in Europe sold their construction and IT businesses and sought to run more tender processes for the supply of services.49

3.4.2 New Energy Market Players in Guangdong Figure 3.8A shows the generator market share in Guangdong among all generation. Figure 3.8B shows the market shares in the liberalised power market. The wholesale market in Guangdong is attracting new players into the market. Of the 13 largest retailers, 3 are privately owned and 10 publicly owned.50 For example, the privately owned Shenzhen Energy Sales and Services Company (SESS) is a new entrant formed (30 January 2015) soon after the market pilot was announced.51 The firm does energy retailing, energy and power contract management, software, renewable  For a discussion in a European context, see Marino et al. (2011).  See Jamasb and Pollitt (2007). 48  See Lohmann (2001). 49  See Hermann and Pond (2012). 50  See Wen (2017). Sung (2017). ‘广东电力市场改革和售电策略探讨’. Guangdong electricity market reform and sale strategy, Mechanical and Electrical Information, 2017 June, 89–90. 51  See the official website of SESS深圳市深电能售电有限公司 (in Chinese) available at: http:// www.sz-sess.com/index 46 47

3  Power Market Reform in China: Lessons from Guangdong 

123

A Guangdong Yudean

24%

China Best Group

33%

Shenzhen Energy Huaneng Group CNOOC

4%

China Resources Group

4%

Datang Group

5%

Zhujiang Investment Corporaon

9%

5% 8%

8%

Others

B

Guangdong Yudean Group

13%

China Resources Power Shenzhen Energy

11%

39%

Shenzhen Energy Sales and Services Huaneng Guangdong Guangzhou Hengyun

8%

Suikai Electricity Guangdong Clear Way

7% 3%

3%

3% 3% 4%

6%

Guangdong Hungfa SGO Power Sales Others

Fig. 3.8  Generator and retailer market shares in Guangdong in 2017. A. Generator capacity shares (Source: http://mp.weixin.qq.com/s/CBdmpsVFppV1j2WskWnPNQ). B. Retailer market shares. (Source: http://www.sohu.com/a/213958006_679911)

124 

M. G. Pollitt et al.

energy projects, building incremental grid and power management research. It focuses on big data and IT management, with a management team drawn from both the electricity and IT sectors. It was the first company to be granted a retail permit. At the centre of its operation is a retail management platform; it has a retail market share of around 10% in the wholesale power market. It can offer a number of value-added services to its retail customers including power system emergency response, technical consultancy, preventative testing, engineering management, price monitoring, load control, accurate measurement in real time and so forth. Many of these services are currently included in the market discount, but eventually SESS may be able to charge separately for some of them. In the future retailers will be able to add value through data mining, focusing on smart energy and smart grids rather than asset heavy solutions. Data mining will allow different consumers in the same sector to be compared to each other, in order to offer better energy efficiency advice. Other market participants include established generation companies, such as China Resources Power (a large conglomerate) who established their CRP Sales company in November 2015; they offer energy and efficiency management services and professional equipment and repair services.52 Retailers are interested in competition in connections, whereby they compete with CSG for network extensions. This is because the profit margin on a grid extension is currently substantially more per TWh than in generation. Interestingly many of the retailers are integrated with generators (e.g. China Resources Power) and with large industrial customers (e.g. SESS53 with BYD54). There are currently three major types of retail contracts that retailers seek to sign with electricity customers.55 First, minimum discount  See the official website of China Resource Power (华润电力) available at: http://www.crpower.com/en/ 53  See China Resources Power (in Chinese) ‘售电公司那么多,谁已经开始行动了?’ available at: http://www.360doc.com/content/16/0514/11/30627394_559016726.shtml 54  See National Business Daily (2015). ‘科陆电子、比亚迪入股民营售电公司’ available at: http://finance.ifeng.com/a/20150327/13586347_0.shtml 55  People’s Government of Guangdong Province (2017a). ‘广东电力市场交易基本规则(试行)’ Retrieved from: http://www.nea.gov.cn/2017-01/20/c_135999956.htm; People’s Government of Guangdong Province (2017c). ‘广东电力市场建设实施方案’, Retrieved from: http://dp. meizhou.gov.cn/show/index/630/15611; See also Shu et al. (2016a, b). 52

3  Power Market Reform in China: Lessons from Guangdong 

125

contracts, where the retailer guarantees a fixed discount on the regulated price and then keeps anything above this that they can save in the wholesale market. Second, sharing contracts where 80–90–95% of the market discount goes to the customers and the retailers keep the rest. Third, combination contracts which combine minimum discounts with sharing. Imbalance charges are imposed if retailers overuse or underuse power relative to their contracted position in the power market. Retailers are incentivised to match supply and demand as they must keep their total imbalance within +/−2% of their contracted amount. Imbalance charges for retailers are set at up to 5% of the power market price. Some retailers share imbalance risk with their customers, while others absorb it up to a point. Retailers can offer value added services through energy efficiency advice and investment. Some retailers demonstrate very high levels of accuracy in matching their contractual position to their actual monthly demand (i.e. +/−3% per individual contract), while others are very inaccurate (i.e. +/−30%). Retailers are not able to trade between themselves but generators can do this from June 2017. An important objective of power market reform in China is to prepare for the internationalisation of the Chinese power sector. The maturing of the Chinese power system will inevitably mean a reduction in the demand for power equipment within China, with consequences for the productive capacity of the power system. One strategic response to this, in line with the ‘One-Belt One-Road’ policy, is that supply chain companies within China will need to seek new markets abroad.56 It further suggests the benefit of spinning out separate companies from existing ones to create more nimble companies better able to bid for contracts in a competitive environment (perhaps, against other Chinese companies). A good example of this is Guangdong Electric Power Design Institute Co. Ltd. (GEDI), a subsidiary of China Energy Engineering Group (mentioned in Chap. 1) which is an engineering and project contracting company,

 ‘China National Development and Reform Commission and National Energy Administration (2016)’, 电力发展 ‘十三五”规划(2016–2020 年)’ NDRC, NEA: Electricity Sector Development under 13th Five-Year Plan (2016–2020), 25 December, 2016, Retrieved from: http://www.sdpc. gov.cn/zcfb/zcfbghwb/201612/P020161222570036010274.pdf 56

126 

M. G. Pollitt et al.

responsible for delivering power projects.57 It was formally separated from China Southern Grid in 2017. GEDI now has around 40% of its business outside China.

3.5 Effects on Operations and Dispatch 3.5.1 International Experience The creation of wide area wholesale markets can affect prices in two ways. First, by exposing the existing prices to challenge it can mean that generators and retailers are forced to set prices which more accurately reflect supply and demand conditions. Thus if there is actually too much generation relative to demand at the initial prices, prices should fall as generators and retailers are forced to reduce prices to bring supply and demand back into balance. This could be as simple as showing up the fact that the previous regulated prices for generation and retail were too high. Thus markets reduce rents (or monopoly power) in the electricity sector (i.e. improve allocative efficiency). Such an effect can be seen in both shorter run (day-ahead) and longer-term (monthly) wholesale markets. However it is important to say that prices might initially be too low relative to supply and demand conditions, as a result of regulation forcing retailers/generators to charge too little. In this case the introduction of a wholesale market will (correctly) raise the price of electricity.58 Second, a wholesale market should bring about an increase in the efficiency of production. This is because whatever the previous system of allocating power between different power plants there are now stronger incentives to allocate power to the least cost power plants first.59 Wholesale markets cause plants to make bids related to costs and to only run if they  See the official website of Guangdong Electric Power Design Institute Co. Ltd. (广东电力设计 院) available at: http://www.cccme.org.cn/shop/cccme8977/index.aspx 58  See Pollitt (2004) for a discussion of the case of Chile where wholesale electricity market prices have been fluctuating up and down (driven by water scarcity) since 1982. Paredes (2003) discusses the links between price fluctuations and performance in the case of Chilean public services. 59  See Newbery and Pollitt (1997) for an analysis of the impact of the introduction of a wholesale power market in England and Wales on operational and investment efficiency. 57

3  Power Market Reform in China: Lessons from Guangdong 

127

are part of the least cost group of plants that can meet system demand. Extension of the market across previously non-integrated areas causes competition between plants on the basis of price bids, where the lower price bid plants will be dispatched first. Because being dispatched now depends on the competitiveness of the price bid in the market, individual plants have strong incentives to cut costs in order to remain competitive. This is especially true of similar plants which are in the price setting part of the generation bid curve. Here, slightly higher costs can be the difference between winning a contract in the wholesale market or not, and in the longer run make the difference  between being viable or being shut down. Thus markets should incentivise plants to minimise costs (i.e. improve productive efficiency). In turn this gives rise to only invest in power plants which run at least cost and which have a positive net present value (NPV) given expected future market prices. This effect of the introduction of wholesale markets would seem to require short-term markets (normally, day-ahead), as it is in efficient real-time operation that potential dispatch savings relative to current operation are likely to be realised. Once again it is important to say that if prices are initially too low relative to the competitive level, a wholesale market will raise them and cause all plants to be paid more and bring forth more generation from higher-­cost plants. The impact of wholesale power markets on underlying efficiency of operation and investment in power generation is a function of prices being allowed to affect the actual dispatch of power plants. In European markets such as Great Britain, exposure to wholesale market prices gives generators an incentive to self-dispatch only their least cost mix of plants. In US markets price bids are used to determine which plants are dispatched by the system operator on the basis of day-ahead price bids, in centrally dispatch systems. In both cases actual dispatch and price bids are closely related. Indeed, in both these types of markets the underlying price bids directly determine the dispatch decisions. Arguments continue as to whether central dispatch or self-dispatch algorithms are to be preferred. Central dispatch economises on the need for individual generation companies to self-optimise and predict what else might be running on the system. Self-dispatch ensures that all information on the firm’s costs and contractual position is taken into account in determining whether it wants to run particular plants, regardless of the payment rules

128 

M. G. Pollitt et al.

of the market. The Competition and Markets Authority in the UK recently concluded60 that GB’s self-dispatch system and the typical central dispatch system in the US were equally efficient. Across Europe and North America market extension has been very important for both allocative and productive efficiency. The efficient use of interconnector capacity between European countries has been a way that production has been reallocated between countries to reduce total system costs, while in PJM market extension has also reallocated production within previously separately dispatched areas. This effectively increases competition within separate markets and ensures that the least cost plants are dispatched across the whole market area.61

3.5.2 Effects on Dispatch in Guangdong There have been substantial impacts on the financial returns to coal-fired power generation in Guangdong as a result of the recent power market reform. Ng (2016) predicts returns to fall from 9% in 2016 to 5% by 2018.62 China Light and Power (CLP) for instance announced a sharp drop in profits in their generation business in southern China, as a result of power market reform.63 Across the CSG area there are four levels of dispatch64: (1) the CSG level, which includes West to East interconnection, (2) provincial grid, (3) city level (including Guangzhou and Shenzhen) and (4) county level, which includes distributed generation (e.g. small hydro). All coal, some gas CHP and all nuclear are subject to provincial-level dispatch in Guangdong. Dispatch decisions by the system operator are not directly influenced by the annual and monthly wholesale market contracts on a real-time basis. Prior to the start of the day-ahead market these are still occurring  CMA (2016, pp. 183–188).  See Mansur and White (2012) for an analysis of benefits of market extension in PJM. 62  Ng (2016). 63  See China Light and Power Group (CLP) (2016). 64  See China Southern Power Grid (CSPG) (2016), Corporate social responsibility report 2016. Retrieved from: http://www.csg.cn/. See also Wang et al. (2006). 60 61

3  Power Market Reform in China: Lessons from Guangdong 

129

according to the dispatch rules which applied before the advent of market trading, that is, on an allocated running hour basis. However the annual and monthly contracts in the power market do influence the allocated running hours over these periods for that portion of power which is in the market. This is different from other power markets where prices in the power market should influence which plants physically run on any given day. Generators are also free to trade power between themselves, to some extent (this was the case before the power market reform). There have been some coal savings as a result of sharper incentives to align supply and demand (in which case it is better to run cheaper coal-fired power plant). Dispatch may be reformed.65 Two models are tested: an auxiliary test where the plant needs to follow instructions as to whether it is to run; and the declaration of plant availability in five days’ time. There are no plans to implement self-dispatch as an answer to incorporate information about underlying contractual position into the power market. Interprovincial trading of electricity is conducted via two trading centres in Guangzhou and Beijing.66 There is a toll fee for power that is transmitted from Yunnan into Guangdong of 0.1147 RMB per kWh (this is very high in relation to the prices paid to generators in Guangdong).67 Guangdong imports one-third of its power, mostly from Yunnan. The amount of trading between Yunnan and Guangdong is subject to negotiation, and while it would seem to be mutually beneficial, there are winners and losers within each province. In Yunnan coal generators will likely see revenues decline (due to competition with coal-fired generation in Guangdong) and market customers will likely pay higher prices, while in Guangdong all generators will likely be worse off, while

 See Ho et al. (2017).  Zhang et al. (2014). 67  See Chen (2018). The 2017 figures for payments to generators in Guangdong are: coal −0.4505 RMB/kWh (source: http://news.bjx.com.cn/html/20170418/820743.shtml), gas −0.745 RMB/ kWh (source: http://shoudian.bjx.com.cn/news/20171010/854086.shtml), wind −0.61 RMB/ kWh (source: http://news.bjx.com.cn/html/20151215/691707-2.shtml) and solar −0.85 RMB/ kWh (source: http://news.bjx.com.cn/html/20161227/799707.shtml) 65 66

130 

M. G. Pollitt et al.

customers are likely better off. Yunnan has a spot market, but it is 95% hydro and not likely to be fully integrated with Guangdong soon.68

3.6 Key Points for Improvement 3.6.1 Discussion of Overall Impressions of Reform One striking thing is the lack of transparency of the composition of the final retail price in different areas of Guangdong. Retailers are unaware of the final prices that their customers actually pay.69 This is because this is still the responsibility of CSG. There is substantial variation in final retail prices for the same type of customer across Guangdong with six major pricing zones70 (originally there were 71) from the point of view of CSG.  County-level final prices are different, particularly between the Pearl River Delta (PRD) and non-PRD areas where different taxes and subsidies apply. There are 21 prefecture-level cities in Guangdong and 19 different prices for electricity. Final prices can vary from 0.1 to 0.2 RMB per kWh on the basis of transmission and distribution charges alone between price areas.71 Final prices are made up of guideline generation prices, utilisation charges, cross-subsidy charges and transmission and distribution charges, in addition to any market discount. The monopoly in transmission and distribution charges is significant in overall reform effects. As discussed in Chap. 2 the introduction of incentive regulation has been a large source of the overall impact of electricity reform on final prices. In January 2017 the NDRC required all  See Guangzhou Power Exchange Center (2017a). 《南方区域跨区跨省月度电力交易规则( 试行)》, Retrieved from: https://www.gzpec.cn/main/index.do. See also Meng, T. Li, C. & Chang, T. (2016). ‘未来广东电力市场下实时调度规则思考’ Guangdong Science & Technology, 2016, 25(16). 69   See Shu et  al. (2016a). ‘基于优化理论市场化的日前电力市场机制设计’ Day ahead Electricity Market Design Based on Market Interpretation of Optimization Theory, Automation of Electric Power Systems. Vol. 40, No. 2. 70  Five pricing zones across Guangdong in addition to Shenzhen. 71  See National Energy Administration China’s official announcement on 13 November 2017 ‘广 东各价区输配电度电价最低-­5.16分/千瓦时’ Retrieved from: http://www.nea.gov.cn/201711/13/c_136749039.htm 68

3  Power Market Reform in China: Lessons from Guangdong 

131

provinces to publish separate electricity network charges.72 Guangdong Provincial NDRC office published its charges in November 2017, fixed in nominal terms for the three years: 2017–2019.73 This office also announced a reduction in the regulated final price of electricity in January 2017 of 0.0233 RMB per kWh for all customers.74 This reduction reflects pressure on CSG to cut funding for some of its T+D projects. Adding this to a discount of −0.0645 RMB/kWh in the bilateral power market means that industrial customers participating in the power market saw reductions of up to −0.0878 RMB/kWh (before payment of the retailer cost) in 2017. This is equivalent to $0.0141/kWh or 10% of the 2015 industrial electricity price (in Table 3.1). Reductions were even higher in 2018 as the annual discount in the bilateral power market increased from −0.0645 to −0.0782 RMB/kWh. However the discount in this market fell in 2019. Three agencies are responsible for regulation of the power sector in Guangdong.75 The Electric Power section of the Economics and Information Commission (EIC) of the Guangdong Development and Reform Commission (DRC) is responsible for the market and licensing of market participants. The Pricing Section of the DRC is responsible for the calculation of T+D charges. The South China Energy Supervision Bureau (part of the NEA) is responsible for some of the monitoring. All three bodies are responsible for monitoring of how competitive the  See NDRC’s official announcement on 22 December 2016b publishing NDRC Price Reform Document 2016 No. 2711 “省级电网输配电价定价办法(试行)” Retrieved from: http://www. ndrc.gov.cn/gzdt/201701/t20170104_834330.html 73  On 7 November 2017, China NDRC just published Guangdong’s transmission and distribution charges for 2017–2019《国家发展改革委关于广东电网2017–2019年输配电价的通知》(发 改价格〔2017〕969号) http://shupeidian.bjx.com.cn/news/20171107/859803.shtml 74  See NDRC Guangdong office’s Price Reform Document 2017 No. 498 ‘广东省发展改 革委关于调整销售电价等有关问题的通知’ Retrieved from: http://www.gddrc.gov.cn/ zwgk/zdlyxxgkzl/djml/xxml/201710/t20171017_433270.shtml See also NDRC Guangdong office’s Price Reform Document 2017 No. 498 ‘广东省发展改革委关于调整销售电价等 有关问题的通知’ Retrieved from: http://www.gzns.gov.cn/zwxxgk/zdlyxxgk/jghsf/201707/ P020170726493877075556.pdf 75  See People’s Government of Guangdong Province (2017b). ‘广东电力市场监管实施办法(试 行)’ Retrieved from: http://www.nea.gov.cn/2017-01/20/c_135999956.htm. See also Sun and Su (2017). China, The Energy Regulation and Markets Review (Edition 6), Edited by David L Schwartz, Law Business Research for a discussion of the government institutions overseeing the energy sector. 72

132 

M. G. Pollitt et al.

market is. There would seem to be a lack of clear responsibility for monitoring competition in the power market between different branches of government. In 2017 there was some legal dispute as to whether China’s Anti-monopoly Act—that is, general competition law—applied to the power sector.76 A new Anti-monopoly Act of 2018 may lead to greater clarity on this point. An independent regulatory agency at the provincial level should be responsible for market participant licensing, market design changes, the setting of regulated network charges and the monitoring of competition. This would have the distinct advantage of pooling administrative resources and experience and in developing regulatory competence on the part of the authorities. Chapter 2 discusses the importance of independent regulation in the international experience of power sector reform and applies this specifically to China. Li and Yu (2017) discuss legal reforms to the supervision system for the power sector which would promote reform77 in a Chinese context. The introduction of new retailers into the power system has had three positive effects. First, it has improved understanding of the nature of the electricity product and customers have been made more conscious of pricing and energy management. Second, the government has gained an understanding of what it means to move from an administered price to a market price. Third, retailers have improved service quality to customers relative to CSG. The ownership of generation in Guangdong is concentrated with the largest company (Yudean, now part of Guangdong Energy Group) having around 35% market share of capacity, with the next largest firm having 20%. Yudean is affiliated with the China Huaneng Group (a national Big five generator),78 but is significantly concentrated within the CSG area. This is true not only of total capacity, but also in terms of peak  The first legal case regarding the applicability of the Act to the power sector appeared in Shanxi in June 2017, when the regulator applied the Act to several generators. This is being challenged by the generators who are challenging whether the Act can apply given that there is not a fully competitive electricity market. See http://finance.sina.com.cn/roll/2017-08-08/doc-ifyitapp2997520. shtml; http://news.xinhuanet.com/2017-06/06/c_1121092318.htm; and https://hk.saowen. com/a/ba4f85af87d5d5a78c9572a60e62b11de7ac497c60818c08dad5a55649576c81 77  Li and Yu (2017). 78  Yudean is 24% owned by China Huaneng Group and 76% by the People’s Government of Guangdong Province (see http://www.gdyd.com/site/yudean/gsjj/index.html). 76

3  Power Market Reform in China: Lessons from Guangdong 

133

generation. This suggests there may be some value in swopping assets between state-owned generators to create more competition in bidding. The current monthly power market only covered 20% of demand in 2017 (though this has increased in 2018, reaching 30% in first half of 2019). This represents 46% of in-province generation (though it is higher for coal-fired power plants) in Guangdong. The marginal cost of power in the power market can be less than the marginal fossil fuel cost of production. This is because of start-up, shut-down and part-loading costs. For a given coal-fired power plant, failure to sell to power in the power market might require reduced power output. If this raises remaining marginal fuel costs (due to part-loading) or requires an expensive shut down of a plant/start-up cost of another plant, then a bid below marginal fuel cost would be optimal. Coal-fired power generation decreased significantly in the first half year of 2019, not only in Guangdong, but also in other provinces, as some coal-fired power plants were closed. A significant early problem with the monthly market is that there was too much demand relative to supply in the partly liberalised market.79 This results in retailers simply bidding the regulated price of power with a slight discount on the final amount of energy that they want to purchase in order to reduce the market price (due to the price determination processes, explained later). The monthly supply and demand curves may not actually cross, as for example appears to have happened in February and March 2017.80 The way the price is determined has changed since the beginning of the power market. In 2016, until February 2017, the price that was paid was determined by a formula which calculated the share of the savings attributable to winning bidders in the auction. This is not an efficient uniform price auction such as we see in most wholesale electricity markets around the world. The calculation in Guangdong was done by  See also Yang et al. (2017). ‘广东电力市场需求侧响应交易机制研究’. Research on Demand Response Trading Mechanism in Guangdong Electricity Market, Guangdong Electric Power, 30(5), 25–34. Pang, P. (2016). ‘电力市场化改革背景下电力需求响应机制与支撑技术’. Electric Power Demand Response Mechanism and Support Technology Under Electric Power Market Reform, Guangdong Electric Power. 80  See Zhang (2017) for a commentary on this figure. 79

A

Electricity (100MWh) 3

4

5

6

7

8

9

0 -100 -200 -300 -400 -500

B 0

2

X axis - Electricity Quantity (100 MWh); Y axis = Relative Price (0.001 Yuan/kWh) 1

2

3

4

5

6

7

8

9 Relative Price (0.001 Yuan/KWh)

Relative Price (0.001 Yuan/KWh)

1

-100 -200 -300 -400 -500

0

1

2

4

3

5

6

7

8

9

-100 -200 -300 -400 -500

System Discount Charge

( 300 )

C

0

0 :

( 1750 )

100

( 2050 )

200

1

:

2

16 0

341.6 683.4

0 :

1

: (1025)

2

700

800

:

410

468.6

16

14

:

14

250

: 5 146.4

: 5

(1025)

Fig. 3.9  Illustrative market clearing and price determination. A. Retailer (green) bids, generator (blue) offers and maximum trading volume (red). B. Calculation of system discount charge. C. Allocation of system discount charge to winning bidders. D.  Calculation of final prices paid to winning retailers and generators. (Source: Jing et al. 2018) (continued)

135

3  Power Market Reform in China: Lessons from Guangdong 

D

Electricity (100MWh)

1 Relative Price (0.001 Yuan/KWh)

0

3

4

5

6

7

8

9

0

-100 -200

2

170.8 234.3

146.4 205

-300

341.7

-400 -500

bid of customers

bid of generators

Clear discount of customers

Clear discount of generators

Fig. 3.9  (continued)

calculating the theoretical savings in both the demand bids and supply offers relative to the undiscounted price (i.e. areas of gross consumer surplus and producer surplus). The sum of these two areas was then used to calculate a system discount charge.81 This was then apportioned 50-50 to the demand and supply side. The demand received total savings equal to half the area, and the supply received prices lower than the regulated price by half the amount of the system discount charge. This ensured that the total discount is matched on the demand and supply side. The savings for demand were apportioned in proportion to their relative absolute discounts. Lower prices for generators were apportioned in proportion to the relative absolute discounts. An example of how the calculation worked is shown in the panels of Fig. 3.9. Thus, in panel A we have three retailers and three generators, who win in the auction. For instance, the least price generator bids to supply 2 units at price of −400. The market clearing price should be −200 and all the retailers in the market should pay this.  With thanks to Phil Chen for explaining this in a note: ‘Clearing mechanism of Guangdong Power Market in 2016’. 81

136 

M. G. Pollitt et al.

In panel B the gross consumer surplus is in orange (300) and the gross producer surplus is in purple (1750). These are summed to give the system discount charge (2050). In panel C the total system discount charge is then divided equally between the winning retailers and the winning generators in proportion to their gross consumer and producer surpluses. For example, the second lowest bidding retailer bids for a consumer surplus of 100, which was one-third of the total winning surplus (= 100/300), and hence receives a discount of one-third of 1025 = 341.6 (the retailer allocation of the system discount charge). In panel D, the allocated system benefit charges are converted into prices to be paid and received. These prices are calculated by dividing the allocated system benefit charges by the number of units demanded or supplied by the winning bidders. In this case the second lowest bidding retailer pays a discounted price of −170.8 (341.6 divided by 2 units). The final result shows that generators receive more than they bid, while retailers pay less than they bid. Importantly intra-marginal bidders affect the outcomes in the market, because the retailer bidding −50 (call it R2) and the generators bidding −400 and −350 (call them G1 and G2) can influence the final prices, even though they are intra-marginal. Thus if R2 had bid −100, it would have changed all of the final prices paid and received, and paid less themselves. Similarly, if G1 had bid −350, this would have changed all of the final prices paid and received, and received more themselves. We show the calculations in the Appendix. This should not happen in a well-designed auction.82 This encourages bidders to game demand bids down and supply bids up, rather than encouraging truthful bidding. Fortunately, this market design was changed. Now if the bids and offers were as in Fig. 3.9A, the market would clear at the mid-point of the marginal generator offer (−200) and the marginal retailer bid (−100), that is, −150. All generators and all retailers receive and pay this price. This does eliminate the potential for intra-marginal generators and retailers to manipulate prices. It is still gameable, as the marginal retailer has an incentive to reduce their bid to further reduce the price, and the fact  On suggestions for good auction design, see Klemperer (2002) and Ausubel and Cramton (2011).

82

3  Power Market Reform in China: Lessons from Guangdong 

137

that market price is a mid-point price, gives rise to the potential for out-­ of-­merit order offerors and bidders to shave their bids because out-turn prices are different from marginal winning offers and bids. A better solution is to realise that the market has surplus supply and that the market clearing price is where the fixed quantity to be sold crosses the supply curve (−200 in this case), that is, the price at which generators are willing to supply the whole market. In 2016 Yunnan was also not using a uniform market clearing price in its auction.83 Instead the highest demand bid and the lowest supply bid were matched and the average of the two is taken and this is the price paid by the demand bidder to the supply bidder.84 This also encourages demands to shave their bids down and suppliers have an incentive to shave their bids up. Guangdong has made significant progress with technical implementation and trialling of its day-ahead market in 2019. The software for implementing a day-ahead 15-minute power market has been written. This calculates PJM type nodal prices at 2000 nodes. This spot market is being implemented by the Guangdong Dispatch Center of CSG and not the Guangdong Power Exchange. Its algorithm will be fully integrated with dispatch. Significantly the market has been trialled, including with real money bids and offers, for the first time, on two days in May 2019 (15–16th). The market delivered significantly lower prices than those seen in the annual and monthly market, for the two particular days (296 and 276 Yuan per MWh). However the trialling has highlighted a number of features of the market which need to be addressed. There has also been progress with the implementation of a day-ahead frequency regulation market. First, there was significant intra-day variation in prices and significant nodal price differentials. The price in certain regions on 16 May was 712 Yuan/MWh. At particular nodes the cap and floor prices (1000 Yuan per MWh and 70 Yuan per MWh) were binding. This did impact the average price and raised the issue of why there was any floor price and how the  Feng (2016a). ‘从云南方案看新电改隐患’ Yunnan local pilot and challenges for China’s recent power sector reform, China Energy, Retrieved from: http://www.cnenergy.org/yw/zc/201602/ t20160205_270260.html 84  See Zhang (2017). See also Tung and Huang (2017). ‘美国 PJM 电力市场及对广东电力改革 的启示’ Experiences in PJM Market in the United States: A Good Reference for the Power Market Reform in Guangdong Power Grid, Yunnan Electric Power, 2017, 45(1). 83

138 

M. G. Pollitt et al.

price cap was chosen (1000 Yuan per MWh is an extremely low price cap). Second, the impact of non-availability of transmission lines was also seen to significantly increase the nodal prices in areas with induced congestion. This illustrated the need for CSG to work with the market in order to minimise the impact of its activity in taking lines out of service for maintenance or system upgrades. Third, with three major generators in the province there is a need to monitor their bidding behaviour closely and regulate bids at certain constrained nodes. The spot market (as of October 2019) has yet to be switched on continuously, amid concerns that prices might both be volatile and ultimately rise above the current annual and monthly contract prices (and indeed above the benchmark regulated industrial price) at some nodes. The plan is that only generators and not customers are exposed to the nodal prices. We discuss in Chap. 4 whether nodal pricing, which gives rise to potentially large price variations across an area in real time, is a necessary feature of power market reform, especially given that power markets in Europe in general have not adopted nodal pricing. Market reform in Guangdong should have implications for nearby provinces. Yunnan has very low retail prices for power.85 A fully functioning power market which included Yunnan and Guangdong would involve hydro generators in Yunnan getting much higher wholesale prices for their power. There are substantial benefits for trading power between other southern provinces in China (e.g. Yunnan and Guizhou, see Zhang et al. 2014).86 This would bid up the wholesale price in the Yunnan power market and potentially increase prices for retail customers in Yunnan. One solution to this would be to identify a ‘hydro benefit’, which would be taxed from hydro producers in Yunnan and used to reduce transmission and distribution charges in Yunnan for connected customers in the province. This was the solution that was implemented in the UK for electricity customers in the North of Scotland following market liberalisation in 1990, continuing a pre-liberalisation benefit from local hydro production.87 This would maintain efficient price signals while ensuring  See Feng (2016a, b) and discussions in Cheng et al. (2018) and Liu et al. (2019).  Zhang et al. (2014). 87  See DECC (2015, p. 9). 85 86

3  Power Market Reform in China: Lessons from Guangdong 

139

that Yunnan customers did not lose out from the negative redistributional effects of market reform. Cross-provincial trading should make use of supply and demand curves for the whole region across which trading is occurring and not be based on arbitrary restrictions on traded quantities, unrelated to available capacity.88 A particular problem is the conflict between the desire for such trading from the central government—who can see its merits—and the individual provinces for whom there will be winners (i.e. electricity customers in Guangdong and electricity generators in Yunnan) and losers (i.e. electricity generators in Guangdong and electricity consumers in Yunnan).

3.6.2 Recommendations for Furthering Reform 1. There is a need to acknowledge that the value of assets in generation will go down with the introduction of a market which reduces prices. If necessary there should be a reallocation of assets between state-­ owned generators to increase competition and spread the value loss. It would also be possible to introduce a competitive transition charge89 on consumers which would collect some of their savings and use them to compensate the generators for losses of asset value directly. 2. In Guangdong, there is a need to complete the move to a day-ahead market for all generation and to integrate this with dispatch. Partial monthly contract markets have successfully encouraged the creation of a new set of market actors—retailers—but they are not generating a proper set of price signals for operation and investment. A complete day-ahead market implies that it is difficult to avoid a big-bang day for trading. This big-bang day approach was the one experienced in the UK and the US wholesale power markets.90 Long-term contracts between generators and customers can be used to hedge financial positions.   Guangzhou Power Exchange Center (2017a). 《南方区域跨区跨省月度电力交易规则 (试行)》, Retrieved from: https://www.gzpec.cn/main/index.do 89  For a discussion in the context of California, see https://www.eia.gov/electricity/policies/legislation/california/assemblybill.html 90  See FERC (2004) for a discussion of this in the context of setting up an independent system operator in the US. Henney (1994) presents the background to the establishment of the power pool in Great Britain on 1 April 1990. 88

140 

M. G. Pollitt et al.

3. It would seem sensible to experiment more fully in one province. A genuine market pilot needs a full set of wholesale electricity markets applied to all generation and demand. A full set of electricity markets should include both markets for energy (yearly, monthly, day-ahead and intra-day) and for ancillary services (particularly frequency and short-term operating reserve). This is not the case, as of 2019, in any existing pilot, including in Guangdong. Guangdong is a good candidate for a comprehensive pilot because of its initially high electricity prices and relatively small electricity sector within its GDP. Continuing the Texas analogy we introduced at the beginning, Texas has gone further with power market reform than any other state in the US.91 The result has been low prices and high renewables penetration. 4. The probability of reversal of power market reform in China seems higher than in many other jurisdictions due to the lack of progress over a 5-year time period and the lack of legislative underpinning of the reform itself. The reform is based on a ruling from the State Council (No.9 of March 2015) which does not have legal force and can be quietly abandoned.92 This is what happened following the 2002 reform due to a combination of rising commodity prices and security of supply concerns. This suggests that there needs to be an additional sense of urgency in reform, which the current process may lack due to the longer political cycle in China (10+ years) of one presidency. A national Electricity Act is in preparation.93 In the UK, the 1987 General Election set a five year maximum (and effectively four year) timetable for power sector reform (which was largely completed by 1991).94 This argues in favour of experimentation to create a workable plan at the provincial level first, AND THEN setting an ambitious time table for reform more generally.  See Adib et al. (2013) for a discussion.  Document No.9 is a policy document, not a law. Some local governments say that sometimes they do not know whether they should just follow policy documents and ignore the content of current Electricity Act (1996). There is therefore an urgent need to revise the current legal frameworks for China’s electric industry in line with the policy goals of the No.9 document. See China Electric Council http://news.bjx.com.cn/html/20161129/792530.shtml 93  Zhang and Dai (2018). See also Wilson, Yang, and Kuang (2015). 94  See Henney (1994, 2011) for a discussion of the reform process in the UK. 91 92

3  Power Market Reform in China: Lessons from Guangdong 

141

 ppendix: How Changing Infra-Marginal Bids A Changes the Auction Results in the 2016 Power Market Auction Design What If R2 (the second and third units of retail demand) had bid −100, instead of −50. Panels A to D change and the R2 receives a bigger discount. See Figs. 3.10 to 3.13 for the calculations. R2’s discount was originally −170.8; it is now −268.8. All other prices are changed. If G1 (the two cheapest units of generation) had bid −350, rather than −400, then they would end up receiving higher payments. Figures 3.14– 3.17 show the calculations. G1’s original payment was a discount of −234.3; it now becomes −206.8. Note all other final prices are changed.

Fig. 3.10  Retailer (green) bids and generator (blue) offers, with maximum trading volume (red)

142 

M. G. Pollitt et al.

Fig. 3.11  Calculation of system discount charge

Fig. 3.12  Allocation of system discount charge to winning bidders

3  Power Market Reform in China: Lessons from Guangdong 

143

Fig. 3.13  Calculation of final prices paid to winning retailers and generators

Fig. 3.14  Retailer (green) bids and generator (blue) offers, with maximum trading volume (red)

144 

M. G. Pollitt et al.

Fig. 3.15  Calculation of system discount charge

Fig. 3.16  Allocation of system discount charge to winning bidders

3  Power Market Reform in China: Lessons from Guangdong 

145

Fig. 3.17  Calculation of final prices paid to winning retailers and generators

References English ACER/CEER (2019). Annual Report on the Results of Monitoring the Internal Electricity and Natural Gas Markets in 2018, Electricity Wholesale Markets Volume November, Slovenia: ACER/CEER. Adib, P., Zarnikau, J., & Baldick, R. (2013). Texas electricity market: Getting better. In F.  P. Sioshansi (Ed.), Evolution of global electricity markets (pp. 265–296). Waltham, MA: Academic Press. An, B., Lin, W., Zhou, A., & Zhou, W. (2015). China’s market-oriented reforms in the energy and environmental sectors. Paper presented at the Pacific Energy Summit. Retrieved from: http://nbr.org/downloads/pdfs/ETA/ PES_2015_workingpaper_AnBo_et_al.pdf Andrews-Speed, P. (2013). Reform postponed. The evolution of China’s electricity markets. In Evolution of global electricity markets. New paradigms, new challenges, new approaches (pp. 531–569). Waltham: Elsevier. Andrews-Speed, P., Dow, S., Oberheitmann, A., Ramsay, B., Smith, V., & Wei, B. (2003). First steps in power sector reform: The case of China’s Guangdong Province. Utilities Policy, 11(3), 169–183.

146 

M. G. Pollitt et al.

APX  – Belpex  – EPEX Spot  – Mercatoelettrico (GME)- Nord Pool Spot  – OMIE OTE (PCR PXs). (2013). EUPHEMIA public description PCR market coupling algorithm. Available at: http://www.apxgroup.com/wp-content/ uploads/Euphemia-public-description-Nov-20131.pdf Ausubel, L., & Cramton, P. (2011). Auction design for wind rights. Report to Bureau of Ocean Energy Management, Regulation and Enforcement. Available at: http://www.boem.gov/uploadedFiles/BOEM/Renewable_ Energy_Program/Regulatory_Information/AusubelCramtonPaper1.pdf Bessembinder, H., & Lemmon, M. L. (2002). Equilibrium pricing and optimal hedging in electricity forward markets. Journal of Finance, 57(3), 1347–1382. Bui, T., Yang, D., Jones, W., & Li, J. (Eds.). (2002). China’s economic powerhouse: Economic reform in Guangdong Province. Basingstoke: Springer. Chau, J., Noria Kia, N., Kilburn, M., & Lorrimar-Shanks, J. (2011). An introduction to Guangdong’s energy market  – Proposed and future solutions. Interconnection paper, Civil Exchange. Chen, R. (2018). Analysis of inter-provincial transfers of supply contracts in South China regional electricity market. mimeo. Cheng, W. (2016). Power retail pilots open in Guangzhou, Chongqing. China Energy Storage Alliance. Retrieved from: http://en.cnesa.org/featured-stories/2016/3/2/power-retail-pilots-open-in-guangzhou-chongqing Cheng, B., Dai, H., Wang, P., Xie, Y., Chen, L., Zhao, D., & Masui, T. (2016). Impacts of low-carbon power policy on carbon mitigation in Guangdong Province, China. Energy Policy, 88, 515–527. Cheng, C., Chen, F., Li, G., Ristic, B., Mirchi, A., Qiyu, T., & Madani, K. (2018). Reform and renewables in China: The architecture of Yunnan’s hydropower dominated electricity market. Renewable and Sustainable Energy Reviews, 94, 662–693. Chick, M. (2007). Electricity and energy policy in Britain, France and the United States since 1945. Cheltenham: Edward Elgar. China Light and Power Group (CLP). (2016). Annual report. Retrieved from: https://www.clpgroup.com/en/investors-information/financial-reports/ our-reports?year=2016 CMA. (2016). Energy market investigation, final report. London: Competition and Markets Authority. DECC. (2015). Hydro Benefit Replacement Scheme & Common Tariff Obligation three year review of statutory schemes: Consultation. London: Department of Energy and Climate Change.

3  Power Market Reform in China: Lessons from Guangdong 

147

FERC. (2004). Staff report on cost ranges for the development and operation of a day one regional transmission organization. Docket no. PL04-16-000. FERC, Washington, DC. Henney, A. (1994). A study of the privatisation of the electricity supply industry in England and Wales. London: EEE Limited. Henney, A. (2011). The British electric industry 1990–2010 the rise and demise of competition. London: EEE Limited. Hermann, C., & Pond, R. (2012). Concentration and disintegration: Company responses in the electricity sector and consequences for employment. In C. Hermann & J. Flecker (Eds.), Privatization of public services: Impacts for employment, working conditions, and service quality in Europe (pp.  33–54). New York: Routledge. Ho, M. S., Wang, Z., & Yu, Z. (2017). China’s power generation dispatch. RFF report. Available at: http://www.rff.org/files/document/file/RFF-RptChinaElectricity.pdf Hurlbut, D., Zhou, E., Bird, L., & Wang, Q. (2017). Transmission challenges and best practices for cost-effective renewable energy delivery across state and provincial boundaries (No. NREL/TP-6A20-67462). NREL (National Renewable Energy Laboratory), Golden, CO. International Carbon Action Partnership (ICAP). (2017). China  – Guangdong pilot system, ETS detailed information. International Carbon Action Partnership. Available at: https://icapcarbonaction.com/en/status-report-2017 International Energy Agency (IEA). (2019). China power system transformation: Assessing the benefits of optimized operations and advanced flexibility options. Paris: IEA Publications. Jamasb, T., & Pollitt, M. (2007). Incentive regulation of electricity distribution networks: Lessons of experience from Britain. Energy Policy, 35(12), 6163–6187. Klemperer, P. (2002). What really matters in auction design. Journal of Economic Perspectives, 16(1), 169–189. Available at: http://people.hss.caltech.edu/~jkg/ what-matters.pdf. Krishna, V. (2010). Auction theory (2nd ed.). Amsterdam: Academic Press. Li, M., & Shen, T. (2016). Guangdong Electric Power Trading Center yesterday officially launched mainly act as a “matchmaker”. Top News. Li, Y., & Yu, J. Q. (2017). The propulsion of China’s electricity-sold side reform. In DEStech transactions on engineering and technology research, (apetc).

148 

M. G. Pollitt et al.

Lin, J., Kahrl, F., Yuan, J., Chen, Q., & Liu, X. (2019). Economic and carbon emission impacts of electricity market transition in China: A case study of Guangdong Province. Applied Energy, 238, 1093–1107. Liu, S., Yang, Q., Cai, H., Yan, M., Zhang, M., Wu, D., & Xie, M. (2019). Market reform of Yunnan electricity in southwestern China: Practice, challenges and implications. Renewable and Sustainable Energy Reviews, 113, 1–24. Lohmann, P. (2001). A ‘hands-on’ or ‘hands-off’ approach in electricity distribution; outsourcing of business processes in a HRM perspective. In 16th international conference and exhibition on electricity distribution, 2001. Part 1: Contributions. CIRED (IEE Conf. Publ No. 482). Mansur, E. T., & White, M. W. (2012). Market organization and efficiency in electricity markets. Retrieved from: http://www.dartmouth.edu/~mansur/ papers/mansur_white_pjmaep.pdf Marino, A., Bertoldi, P., Rezessy, S., & Boza-Kiss, B. (2011). A snapshot of the European energy service market in 2010 and policy recommendations to foster a further market development. Energy Policy, 39(10), 6190–6198. Newbery, D. M., & Pollitt, M. G. (1997). The restructuring and privatization of Britain’s CEGB: Was it worth it? Journal of Industrial Economics, 45(3), 269–303. Ng, E. (2016). Chinese power generators face tougher times as deregulation sparks competition. South China Morning Post. Ofgem. (2016). Wholesale energy markets in 2016. London: Ofgem. Ofgem. (2017). State of the energy market 2017 report. London: Ofgem. Paredes, R. (2003). Redistributive impact of privatization and the regulation of utilities in Chile. In C. Ugaz & C. Waddams Price (Eds.), Utility privatisation and regulation – A fair deal for consumers? Aldershot: Edward Elgar. Pike, L., & Zhe, Y. (2017). Five Things to Know About China’s National Carbon Market. China Dialogue, December 19. Pollitt, M. (2004). Electricity reform in Chile: Lessons for developing countries. Journal of Network Industries, 5(3–4), 221–262. Regulatory Assistance Project (RAP). (2016). Wholesale electricity markets and pricing in China: How is reform going? Available at: http://www.theenergycollective.com/raponline/2389923/wholesale-electricity-markets-and-pricingin-china-how-is-reform-going-2 Reinhold Cohen. (2015). China’s new intellectual property courts. Reinhold Cohen IP Attorneys newsletters. Available at: http://www.rcip.co.il/en/article/chinas-new-intellectual-property-courts/

3  Power Market Reform in China: Lessons from Guangdong 

149

Stoft, S. (2002). Power system economics designing markets for electricity. Piscataway: IEEE Press. Sun, M., & Su, H. (2017). China. In D. L. Schwartz (Ed.), The energy regulation and markets review (6th ed.). London: Law Business Research. Taylor, S. (2007). Privatisation and financial collapse in the nuclear industry: The origins and causes of the British energy crisis of 2002. Abingdon: Routledge. Wang, C., Chen, H. R., Pan, W. L., & Xu, Z. (2006). Reliability and economic evaluation of the China southern power grid. In Power systems conference and exposition, 2006. PSCE’06. 2006 IEEE PES (pp. 854–859). IEEE. Wang, W., Luo, Y., Xie, P., Luo, Z., & Zhao, D. (2016). The key elements analysis of Guangdong & Shenzhen ETS & tips for China national ETS construction. Chinese Journal of Population Resources and Environment, 14(4), 282–291. Wilson, S., Yang, Y., & Kuang, J. (2015). China’s electricity sector. In Ligang Song, Ross Garnaut, Cai Fang & Lauren Johnston (Eds.) China’s Domestic Transformation in a Global Context, Canberra: ANU Press, p.175–210. Xinhua Finance. (2015). Shenzhen given nod to pilot new power transmission. Available at: http://en.xfafinance.com/html/Industries/Utilities/ 2015/40163.shtml Yu, D., Qiu, H., Yuan, X., Li, Y., Shao, C., Lin, Y., & Ding, Y. (2017). Roadmap of retail electricity market reform in China: Assisting in mitigating wind energy curtailment. IOP Conference Series: Earth and Environmental Science, 52(1), 012031. Zeng, L., Zhang, C., Chen, L., & Xiang, X. (2004). Guangdong electric power market reform: Options and impact. Working paper #33, program on energy and sustainable development, Stanford University. Zhang, H. (2017). China power market: Too young, too irrational. Retrieved from: https://www.linkedin.com/pulse/china-power-market-too-young-irrational-huiting-zhang/ Zhang, D., & Dai, H. (2018). Financing China’s electricity sector. In L. Lester & M. Thomas (Eds.), China’s electricity sector. Singapore: Palgrave Pivot. Zhang, S. L., Sun, Y. M., Jing, Z. X., & He, Z. X. (2014). Research on transfer and replacement mechanism design of inter-provincial electricity treatment in South China electricity market. Applied Mechanics and Materials, 494, 1695–1701.

150 

M. G. Pollitt et al.

Chinese China Electric Council. (2017). “告别无序竞争 电力市场呼吁售电专业化 时代” Say goodbye to the unregulated competition. http://www.cec.org. cn/xinwenpingxi/2017-10-30/174433.html China National Development and Reform Commission. (2015a). “国家发 展改革委关于深圳市开展输配电价改革试点” 发改价格【2014】 2379号. Retrieved from: http://jgs.ndrc.gov.cn/zcfg/201411/t20141104_ 639639.html China National Development and Reform Commission. (2015b). “电力体制 改革解读” Analysis of electricity institutional reform. Beijing: Remin Publisher. China National Development and Reform Commission. (2016a). “国家发展 改革委 国家能源局关于印发《电力中长期交易基本规则(暂行)》” 发改能源〔2016〕2784号. Retrieved from: http://www.ndrc.gov.cn/zcfb/ zcfbtz/201701/W020170112319053238252.pdf China National Development and Reform Commission. (2016b, December 22). “省级电网输配电价定价办法(试行)” NDRC price reform document 2016 no. 2711. Retrieved from: http://www.ndrc.gov.cn/gzdt/201701/ t20170104_834330.html China National Development and Reform Commission. (2017). “国家发展改 革委关于印发《省级电网输配电价定价办法(试行)》” 发改价格 〔2016〕2711号. Retrieved from: http://www.ndrc.gov.cn/fzgggz/jggl/ zcfg/201701/t20170104_834333.html China National Development and Reform Commission and National Energy Administration. (2016, December 25). “电力发展“十三五”规划 (2016–2020 年)” NDRC, NEA: Electricity sector development under 13th five-year plan (2016–2020). Retrieved from: http://www.sdpc.gov.cn/zcfb/ zcfbghwb/201612/P020161222570036010274.pdf. China National Development and Reform Commission Guangdong. (2017). “ 广东省发展改革委关于调整销售电价等有关问题的通知” Price reform document 2017 no. 498. Retrieved from: http://www.gzns.gov.cn/ zwxxgk/zdlyxxgk/jghsf/201707/P020170726493877075556.pdf China National Energy Administration. (2017, November 13). “广东各价区 输配电度电价最低-­5.16分/千瓦时”. Retrieved from: http://www.nea. gov.cn/2017-11/13/c_136749039.htm China Resources Power. (2016). “售电公司那么多,谁已经开始行动了?” Available at: http://www.360doc.com/content/16/0514/11/30627394_ 559016726.shtml

3  Power Market Reform in China: Lessons from Guangdong 

151

China Southern Power Grid (CSPG). (2016). Corporate social responsibility report 2016. Retrieved from: http://www.csg.cn/ China5e Research Centre. (2016). “中国电改试点进展政策研究与建议” The policy study of China’s power sector reform pilot project development and recommendations. China5e research centre report. Beijing. Feng, Y.-C. (2016a). “从云南方案看新电改隐患” Yunnan local pilot and challenges for China’s recent power sector reform. China Energy. Retrieved from: http://www.cnenergy.org/yw/zc/201602/t20160205_270260.html Feng, Y.-C. (2016b). “如何理解中国电力体制改革:市场化与制度背景” How to understand China’s power sector reform: Market reforms and institutional context. Retrieved from: http://www.chinareform.org.cn/economy/ price/refer/201611/t20161130_25845 Guangzhou Power Exchange Center. (2017a). 《南方区域跨区跨省月度电 力交易规则(试行)》. Retrieved from: https://www.gzpec.cn/main/ index.do Guangzhou Power Exchange Center. (2017b). “广东电力市场2017年半年 报告” 2017 Guangdong electricity market report (January to June). Retrieved from: https://www.gzpec.cn/main/index.do Jing, Z., Chen, Z., & Zhu, J. (2018). “中国电力市场月度竞价机制介绍与 分析” Introduction and analysis on monthly centralized bidding rules of power market in China, walking into the electricity market 走进电力市场. South China University of Technology (SCUT). Liu, D. (2017). “电力市场、碳排放权市场和绿色证书市场的协调发 展.”Electricity market, carbon market and green certificate mechanisms development, Industry perspective, China Electrical Equipment Industry, 2017.07. Meng, T. Li, C. & Chang, T. (2016). “未来广东电力市场下实时调度规则 思考” GuangDong Science & Technology, 2016, 25(16). National Business Daily. (2015). “科陆电子、比亚迪入股民营售电公司”. Available at: http://finance.ifeng.com/a/20150327/13586347_0.shtml Pang, P. (2016). “电力市场化改革背景下电力需求响应机制与支撑技术.” Electric power demand response mechanism and support technology under electric power market reform. Guangdong Electric Power, 29(1), 70–78. People’s Government of Guangdong Province. (2017a). “广东电力市场交易 基本规则(试行)”. Retrieved from: http://www.nea.gov.cn/201701/20/c_135999956.htm People’s Government of Guangdong Province. (2017b). “广东电力市场监管 实施办法(试行)”. Retrieved from: http://www.nea.gov.cn/201701/20/c_135999956.htm

152 

M. G. Pollitt et al.

People’s Government of Guangdong Province. (2017c). “广东电力市场建设 实施方案”. Retrieved from: http://dp.meizhou.gov.cn/show/ index/630/15611 Shu, C., Zhong, H., & Xia, Q. (2016a). “基于优化理论市场化的日前电力 市场机制设计”Day ahead electricity market design based on market inter-­ pretation of optimization theory.  Automation of Electric Power Systems, 40(2), 9–21. Shu, C., Zhong, H., Xia, Q., Tan, Y., Hu, J., Yeo, X., & Wang, Y. (2016b). Monthly elasticity of the electricity market mechanism design. Proceedings of the CSEE, 36(3), 587–595. South China Energy Regulatory Bureau of National Energy Administration of PRC China. (2015). “广东省电力大用户与发电企业直接交易扩大试 点工作方案”. Retrieved from: http://nfj.nea.gov.cn/adminContent/initViewContent.do?pk=zwgk1492 Sung, Y. (2017). “广东电力市场改革和售电策略探讨.” Guangdong electricity market reform and sale strategy,  Mechanical and Electrical Information, 2017 June, 89–90. Tsai, et al. (2017). “英国电改专家怎么看待中国电改,给了什么建议? —— 专访剑桥大学能源政策研究所副所长迈克尔·波利特教授” Interview with Professor Michael Pollitt. Energy Observer. Retrieved from: http://www. eothinker.com/eo/show.php?itemid=569 Tung, C. & Huang, S. (2017). “美国 PJM 电力市场及对广东电力改革的启 示”Experiences in PJM Market in the United States: A Good Reference for the Power Market Reform in Guangdong Power Grid, Yunnan Electric Power, 45(1), 32–35. Wen, H. (2017, February). “广东试轨” Guangdong pilot. Energy Observer. Yang, W., Tseng, C., Wang, F., & Guo, M. (2017). “广东电力市场需求侧响 应交易机制研究.”Research on demand response trading mechanism in Guangdong electricity market. Guangdong Electric Power, 30(5), 25–34. Zhang, et  al. (2015). “售电侧市场放开国际经验及其启示” International experience and lessons in power sales side market liberalization. State Grid Energy Research Institute. Retrieved from: http://www.aeps-info.com/aeps/ ch/reader/create_pdf.aspx?file_no=20151128001&flag=1& journal_id=aeps&year_id=2016 Zhang Lizi, & Xu Chuanlong. (2017). “中国能源大省电力市场建设经验分 析.” Electricity market construction experience of major provinces of energy in China. Electric Power, 50(4), 7–10.

4 How Industrial Electricity Prices Are Determined in a Reformed Power Market: Lessons from Great Britain for China In collaboration with Lewis Dale

4.1 Introduction This chapter aims to unpack how the price of industrial electricity is determined within the liberalised power market in Great Britain in the context of the ongoing reform of the Chinese power sector, initiated by the March 2015 No. 9 document. It is organised around a discussion of the various components of the final industrial price of electricity (namely the wholesale price, the retail margin, network charges, the costs of system operation and government taxes and levies). In the case of each component, we consider what drives its total cost and its pricing structure. In this chapter, we begin by discussing the components of the price of industrial electricity in Great Britain, as an example of a fully reformed electricity market, where the market is roughly comparable in size to a reasonably large Chinese province. We proceed to discuss the key actors in the liberalised electricity system in Great Britain, before unpacking each of the components of the price. We discuss the market-determined elements first, then go on to introduce and discuss the regulated elements of the price before finishing with the central government-determined © The Author(s) 2020 M. G. Pollitt, Reforming the Chinese Electricity Supply Sector, https://doi.org/10.1007/978-3-030-39462-2_4

153

154 

M. G. Pollitt

price components. Our discussion covers the determination of the wholesale price, the retail margin, transmission charges, system balancing charges, distribution charges and environmental levies and taxes. In each of these cases we discuss the process by which they are determined (led by the market, the regulator, the central government or more than one) and the specific lessons for China. We conclude by emphasising some of the high-level lessons on electricity price determination for China.

4.2 H  ow Is the Industrial Electricity Price Set in Great Britain1 The final retail price of industrial electricity is made up of six elements in Great Britain. These are a combination of unregulated market-­determined elements (the wholesale price and the retail margin); regulated charges (transmission and distribution charges); central government-determined levies and taxes; and mixed elements (system balancing charges) that are made up of both regulated and market-determined costs. The final price (charged by retailers) is not regulated for typical industrial customers. Regulated charges are determined by an independent regulatory agency. Central government levies and taxes are the responsibility of the Finance Ministry (HM Treasury in the UK). In the UK we distinguish between industrial, commercial and residential users, so the category of industrial users is narrower than in China, where ‘industrial’ covers both industrial and commercial customers. The breakdown of the final price of industrial electricity in the UK in 2016 is shown in Table  4.1, for industry consuming more than 2000 MWh per year. This shows the underlying sources of the components that make up the price. It does not necessarily correspond to the revenues collected by different parties within the system. Thus, for example, the actual price of wholesale electricity will include the cost of carbon pricing, as this is payable by fossil fuel-based electricity generators.  For electrical purposes, the  UK is divided up  into two administrative areas: Great Britain and Northern Ireland. Electricity in Great Britain is regulated by Ofgem. Some of the statistics we refer to below include Northern Ireland in the UK, but Great Britain is responsible for around 98% of electricity consumption in the UK. 1

155

4  How Industrial Electricity Prices Are Determined…  Table 4.1  Breakdown of industrial price in the UK 2016 Market-determined prices Generation costs Retailer (supplier) costs Regulated charges Transmission charges System balancing charges Distribution charges Levies and taxes Renewables obligation FITs Hydro-Benefit Scheme Climate change levy Carbon reduction commitment Carbon pricing

Euro/MWh 39 7.15 11 2.6 9 17.5 5.24 0.2 2.66 4.4 16.25 115

% 40.1 33.9 6.2 19.7 9.6 2.3 7.8 40.2 15.2 4.6 0.2 2.3 3.8 14.1 100

Source: Derived from Grubb and Drummond (2018). Before discounts Band 1D to 1F customers, 2000 MWh+ annual demand. £1 = 1.16 Euros

Generators’ revenue will also include their transmission charges. The breakdown shows that roughly 40% of the cost is market determined, 20% is determined by the regulator and 40% is determined by the central government. We will discuss each of the sub-elements in the chapter. Note the price of industrial electricity in the UK in 2016 is roughly equal to the price of industrial electricity in Guangdong in 2015 (Chap. 3). At the outset, it is important to clarify the role of government in the UK electricity sector: the government does not determine the final price2; the regulator does determine the maximum revenue for regulated elements; the regulator does approve the tariff methodology for regulated charges; the regulator does determine security of supply requirements and penalties; the government does monitor competition in power markets. While electricity market reform may have the aim of delivering low prices, lower prices are not always the right answer. When fossil fuels are becoming more expensive or generating capacity is getting scarce, higher prices may be the right answer.  In the residential sector, there is retail price cap (known as ‘safeguard tariff’). However, this is a maximum tariff. Actual tariffs can be lower than this. 2

156 

M. G. Pollitt

4.3 T  he Key Actors in the Electricity System in Great Britain The UK electricity sector in 2017 can be characterised as follows.3 It had final consumption of 301 TWh, this has been falling since its peak in 2005 (it was 14% lower in 2017 than the 2005 peak). This reflects changing industrial structure and a large rise in energy efficiency (household energy demand is 16% below its peak, in spite of a rise in the number of households). Energy supply in 2017 was 353 TWh, of which 4.2% was imports. Of domestic generation of 336 TWh, 40.4% was from natural gas, 29.3% was renewables and 20.6% was from nuclear, with only 6.7% from coal. The speed and extent of the transition away from coal and towards renewables in the UK electricity system has been remarkable and demonstrates—in line with more general evidence4—that electricity reform is not incompatible with ambitious environmental targets. In 2010 renewables made up 6.9% of electricity generation, while coal was 28%5; seven years later this has been more than completely reversed. In 1990, at the time of the introduction of the electricity market, the UK was generating 72% of its electricity from coal.6 The reform of the electricity sector in the UK from 1990 significantly changed the institutions involved in the electricity sector.7 In England and Wales the monopoly public generation and transmission company (the CEGB) was broken up and 12 regional electricity distribution and retail companies were able to enter generation and compete with one another for retail customers (via legally separate retail businesses). In  See Digest of United Kingdom Energy Statistics 2018, (BEIS 2018).  See, for example, Vona and Nicolli (2014). 5  See DECC (2011). 6  See DECC (2009). 7  See Henney (1994) for the definitive discussion of what happened (and why) at the time of privatisation to the structure of the electricity industry. For an excellent summary of the GB experience following privatisation, see Newbery (2000, 2005). For a discussion of the electricity privatisation in the context of the general privatisation programme in the UK, see Pollitt (1999), and for a discussion of electricity liberalisation in the global context of energy market liberalisation, see Pollitt (2012a). 3 4

4  How Industrial Electricity Prices Are Determined…  Great Britain (England & Wales and Scotland)

Interconnctors

Tx

EDF

Centrica

Iberdrola SSE

IPPs

National Grid Electricity Transmission

Scottish Power / SSE NIE

Dist.

Supply

Northern Ireland

RWE

EON

Gen.

157

14 Regional Distribution Cos. (6 owners) EON

RWE

EDF

Centrica

Iberdrola

SSE

IPPs

6 principal supply companies Many independent suppliers (vertically integrated with generators) (generally not integrated with generators)

Fig. 4.1  The structure of the electricity industry in the UK

Scotland and Northern Ireland incumbent integrated generation, network and retail companies were also subject to breakup and competition. In mid-2018 there were 170 licensed generators and 64 licensed non-­ domestic electricity retailers in Great Britain.8 Figure 4.1 shows the current structure of the industry. In Great Britain generation and retail is dominated by ‘big’ six generator-retailers, which have arisen from incumbent companies. Transmission in England and Wales is owned by National Grid and distribution is owned by six companies (including UKPN and WPD). National Grid is the system operator for the whole of Great Britain. While some companies continue to have interests in generation, distribution and retail, this is now much less common and there is strict legal unbundling of distribution network businesses from the competitive parts of the electricity sector. Independent power projects and new retailers have taken significant market shares from the ‘big’ six companies, and the generator and retailer market shares are not matched within the ‘big’ six.   Source: Ofgem (2018a), https://www.ofgem.gov.uk/system/files/docs/2018/10/state_of_the_ energy_market_report_2018_1.pdf 8

158 

M. G. Pollitt

EdF 24%

Others 30%

trica

Cen

Un ipe

r4

%

RWE 14%

5%

er h Pow Scottis 4% % G4 EC

SSE 8% Drax 7%

Fig. 4.2  Wholesale market shares 2017. (Source: Ofgem (2018a), State of the Energy Market Report 2018, p. 50)

Others 14%

EdF 19%

Total Gas and Power 6% Smartest Energy 7%

npower 17%

Engie 7% SSE 10% Drax Group 10%

E.ON 10%

Fig. 4.3  Business retail market shares 2017. (Source: Large-scale electricity profile class 5–8 +HH: Ofgem (2018a), State of the Energy Market Report 2018, p. 38)

Figures 4.2 and 4.3 show the market shares in both the wholesale electricity market and business retail market in 2017. The HHI index for the wholesale electricity market is around 1034, indicating that market

4  How Industrial Electricity Prices Are Determined… 

159

concentration is relatively low, equivalent to ten equally sized firms competing with each other. For the retail market the HHI is even lower, at around 1000. There are 7 electricity distribution network companies in the UK (UKPN, WPD, Northern Powergrid, Electricity Northwest, SP Energy Networks, Scottish and Southern Energy [SSE] Power Distribution and Northern Ireland Electricity [NIE]), covering 15 monopoly distribution areas. Of these groups only two are now part of companies with either generation or retail interests in the UK (SP Energy Networks and SSE Power Distribution). There are four onshore transmission companies in the UK (National Grid, SP Energy Networks, SSE Power Distribution and NIE). The ownership structure (as of mid 2019) of the electricity industry in Great Britain is very diverse. Two of the ‘big’ six generator retailers—SSE and Centrica—are listed on the London Stock Exchange, together with the generator Drax and the transmission company National Grid. The other four major generator retailers are subsidiaries of large pan-­European energy companies—EdF, RWE, E.ON, Iberdrola. Of the distribution companies UKPN is owned by Hong Kong investors, while WPD is owned by PPL, a listed energy company in the US, Northern Powergrid is owned by Berkshire Hathaway and Electricity Northwest is privately held. The ownership of the UK electricity industry reflects significant foreign ownership. The attraction of holding UK electricity assets reflects diversification by shareholders in other countries. This makes UK domestic regulation easier (because it weakens the lobbying power of producer interests and enables the regulator to focus on UK consumers). It also facilitates reciprocal UK investment abroad (not just in the electricity sector). Other key actors in the electricity sector in Great Britain are worth mentioning. Elexon administers the balancing and settlement system for reconciling payments between generators and retailers within the electricity industry. This is an extremely important part of any restructured industry. Elexon describes itself in the following way: ‘In March 2001, the Balancing and Settlement Code (BSC), was launched as part of NETA (New Electricity Trading arrangements). ELEXON administers the Code on behalf of the UK electricity industry. We provide and procure the

160 

M. G. Pollitt

services needed to implement the code and compare how much electricity generators and suppliers say they will produce or consume with actual volumes. We work out a price for the difference and then transfer funds.’9 Elexon was established by an obligation in National Grid’s transmission licence and, technically, is wholly owned by National Grid but operates at arm’s length to it with a completely separate governance structure. The power market is also characterised by a number of other players. These include aggregators (who aggregate up both generation and demand on behalf of smaller generators and smaller non-domestic customers). There are around 19 of these.10 Traders trade electricity financially on the available power exchanges. Power can be traded across multiple platforms a day ahead (or under longer-term contracts). These trading platforms include APX Power UK and N2EX.  There are around 75 traders on APX.11 Interconnectors are also significant in the Great Britain market. These offer the ability to supply power both into and out of GB (acting as generators and loads). Currently these interconnectors are: 2 GW to France (IFA); 1 GW to the Netherlands (BritNed); 500 MW to Northern Ireland (Moyle) and 500  MW to the Republic of Ireland (East West). There is another 3.4 GW under construction (as of October 2018). Finally, it is important to note the role of the regulators and the government within the electricity system. While there is no state ownership of operational electricity assets12 in the UK, the government does influence the industry via the regulatory regime and central government ministries. The GB industry is directly overseen by two independent regulators: Ofgem (the Office of Gas and Electricity Markets) and the CMA (Competition and Markets Authority) which is the general competition authority. The primary duty of these two regulators is to oversee the competitiveness of the electricity sector in both the wholesale and retail markets and to approve and monitor monopoly network charges. These  https://www.elexon.co.uk/about/  See PA Consulting Group (2016), https://www.ofgem.gov.uk/system/files/docs/2016/07/aggregators_barriers_and_external_impacts_a_report_by_pa_consulting_0.pdf 11  See http://www.epexspot.com/en/membership/list_of_members 12  The government does own decommissioning nuclear power plants, and some test reactors. Municipalities have limited interests in local electricity companies, for example, Bristol Energy. 9

10

4  How Industrial Electricity Prices Are Determined… 

161

two agencies operate at arm’s length from the government. However, the central government, led by the Department for Business, Energy and Industrial Strategy (BEIS), sets regulatory framework, can issue guidance to the regulator, sets the subsidy and tax regime and can refer the whole industry for investigation to the CMA.13 Members of the regulatory boards are appointed by the government for a fixed term.

4.4 Wholesale Prices We now turn to the elements of the final price of industrial electricity and how these are determined. An obvious place to start in thinking about the wholesale electricity price element of the final price is the spot market price for electricity. So, what do we mean by a spot market in a liberalised power market? We normally mean the main near real-time market (see Stoft 2002) in which the wholesale prices determined every hour, half-hour, 15 minutes or 5 minutes from supply offers and demand bids from individual generators and retailers wishing to sell and buy power. Underlying bids and offers guide dispatch of individual power plants. In many power markets generators can declare prices (e.g. in PJM market in the US) at which they are willing to be dispatched or quantities (e.g. in the market in GB) which they want to be dispatched to system operator. ‘Spot’ prices/quantities should reflect the underlying value of generation and loads. The system operator uses such spot prices/quantities to dispatch the system paying attention to the need for instantaneous correction to match supply and demand in real time, on the basis of constraints and balancing markets/contracts. There is always an issue of how to link spot market prices (which are often day-ahead) and physical dispatch. Instantaneous bidding is neither possible nor desirable because instantaneous prices cannot change behaviour in real time and might lack transparency (as in practice they have to be calculated ex ante).  See CMA (2017), https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/ file/624706/cma3-markets-supplemental-guidance-updated-june-2017.pdf 13

162 

M. G. Pollitt

Within a power market actual real-time grid stability is physically maintained in much the same way as in traditional vertically integrated power systems, following ‘gate closure’ (the last chance for generators and loads to change their market position). However spot markets do create opportunities for withholding by generators as in the California electricity crisis of 2000–01,14 whereby generators manipulated market prices by withdrawing some capacity deliberately to drive up prices on their remaining generation. This sort of behaviour must be monitored, detected and penalised. A crucial part of the wholesale market is the presence of retailers who buy power on behalf of their customers. In Great Britain retailers need to buy wholesale power to cover 100% of their power sales. Retailers are often integrated with generators. Retailers can use spot and forward markets and bilateral contracts. Bilateral contracts are traded in OTC (over-­ the-­ counter) markets directly between generators and suppliers.15 Derivative energy products are available.16 A lot of power is bought and sold on bilateral contracts for 12–18 months, often linked to spot prices. Retailers often hedge using rolling contracts (e.g. by rolling over 18-month contracts every month, thus taking 18 months for any sustained adjustment in the underlying prices to be fully reflected in their wholesale generation contracts). Smaller retailers tend to use shorter-term contracts. This is because there is more relative uncertainty for them on their future demand and they compete on the basis of short-term price competitiveness rather than reputation. Some jurisdictions (e.g. in South America) specify the nature of contracts that should be entered into for regulated retail customers. This can be done indirectly by specifying the benchmark wholesale contract price that will be used in the calculation of the maximum regulated retail price. In GB the bilateral contract markets and the power exchange (PX) are not run by the system operator (SO). Instead the SO operates the half-­ hourly balancing market and other ancillary services markets (e.g. for frequency response). Ninety-seven per cent of all wholesale energy is  See Sweeney (2002).  For a discussion of the operation of this market, see CMA (2016a). 16  See London Energy Brokers Association: www.leba.org.uk 14 15

4  How Industrial Electricity Prices Are Determined… 

163

self-­dispatched in GB. This is different from a compulsory pool/mandatory day-ahead (DA) market such as used in the US (e.g. by PJM) where the system operator uses the day-ahead market to guide pricing and dispatch in real time.17 How self-dispatch works is as follows. Generators can give their Final Physical Notification (FPN) to the system operator of which plants they want to run up to one hour ahead of real time, that is, generators can adjust their stated position up to that time. After that, the system operator takes control of the plants with the object of minimising the costs of any balancing actions (i.e. the system operator is the sole counterparty to balancing market transactions). Production (generator) and consumption (retailer) accounts transactions (Energy Contract Volume Notifications [ECVN]) must also be declared with consumption accounts noting position in all 14 distribution areas (as separate BMUs). Generators and retailers have a strong incentive to self-balance their individual positions either physically or via their participation in the balancing mechanism. Failure to self-balance implies exposure to balancing charges arising from the system operator’s need to buy or sell power to exactly balance the system. Balancing charges are thus calculated to incentivise balance and encourage accurate supply and demand matching. There is an incentive to bid accurately as under competition law there is an up to 10% of turnover fine possible for market power abuse. Such self-dispatch is different from PJM or the Guangdong spot market (discussed in Chap. 3), but with sophisticated players this allows generators to reflect all of their internal costs in their notifications to the system operator. Self-dispatch offers the potential for improved market position/performance with plant risk issues in ways not facilitated by central dispatch algorithms, which although they can take formal account of many plant characteristics (such as ramping costs) cannot take into account all of the costs faced by an individual unit in being dispatched. It is important to note that the efficiency of self- versus central dispatch is usually measured with respect to a central dispatch calculation; thus it  See for example: http://www.gridscientific.com/images/electricity_trading_arrangements_beginners_guide.pdf, https://www.elexon.co.uk/wp-content/uploads/2017/11/BSCP01_v17.0.pdf and Onaiwu (2009). 17

164 

M. G. Pollitt

is biased towards finding central dispatch more efficient and not modelling hidden constraints in system.18 Power plants in Great Britain are dispatched in real time by National Grid; there are no other dispatch layers. The regional distribution companies do not dispatch generation or loads, though they may occasionally exercise constraint management contracts within the distribution system. Larger generators (on the distribution system) must be visible to the transmission system operator (TSO) above 50  MW (as a balancing mechanism unit—BMU) and can register below this size threshold.19 Interconnectors to France, Ireland and the Netherlands act like generators and loads. However, there is some inter-TSO collaboration through CORESO. CORESO is a Regional Security Coordinator (RSC) of the EU single electricity market which brings together seven national system operators in western Europe which does coordinate ‘from a few days ahead until Intraday (few hours before real time)’.20 The wholesale power market is closely linked to fossil fuel markets and carbon markets, via the fact that fossil fuel prices and carbon prices are an important cost component for any oil-, gas- or coal-fired generation unit. Carbon pricing (both emissions permits and taxes) appears in wholesale generator costs as an extra fuel cost as fossil fuel generators have to ‘burn’ carbon permits and pay carbon taxes when they generate and produce carbon dioxide. The figures for wholesale costs in Table 4.1 net off the impact of carbon pricing, but reported wholesale prices include these costs. We discuss their determination in a separate section below. In Great Britain, most subsidised low carbon generation participates in the energy market like any other generator in real time and receives top-up revenue from contract for difference (CFD) and renewable obligation certificate (ROC) payments. Small FIT (feed in tariff) generators (mostly households) receive payment from the retailer that they are contracted to sell to (the so-called FIT Licensee).21  See for example Sioshansi et al. (2008).   See National Grid (2011), https://www.nationalgrid.com/sites/default/files/documents/ Managing%20Intermittent%20and%20Inflexible%20Generation%20in%20the%20 Balancing%20Mechanism%20Consultation.pdf 20  See https://www.coreso.eu/ 21  See https://www.ofgem.gov.uk/environmental-programmes/fit/about-fit-scheme 18 19

4  How Industrial Electricity Prices Are Determined… 

165

While wholesale power markets are not very concentrated in the UK now, that has not always been the case and there have been many examples of market power problems in electricity wholesale markets.22 Indeed, tacit collusion is a very real problem in electricity markets at the wholesale and retail market levels due to repeated interaction of companies making bids and offers many times a day within transparent power markets. The regulator, Ofgem, has done a lot on transparency of profitability.23 The GB market currently falls under REMIT legislation on energy market integrity and transparency.24 Ofgem has referred the whole market to the CMA recently who examined competition thoroughly over the period 2014–16.25 Alex Henney (2011) documents the eight-year process that followed the introduction of the competitive wholesale power market in Great Britain, that eventually led to the much more competitive market that we see today. This consisted of a series of investigations and reports with recommendations. These included the Pool Price Enquiry December 1991 (by Offer—the then electricity only regulator which predated Ofgem) which found gaming by the two incumbent fossil fuel-generating companies declaring plant unavailable day ahead to drive up capacity price on the day and the Report on Constrained-On Plant Oct 1992 (by Offer) which found gaming by behind constraint plants. Subsequent investigations led to significant structural changes: the Pool price statement July 1993 (Offer) allowed the largest customers to bid in the Pool (to reduce generator market power) and then following the threat to refer the two largest generators to the competition authority in February 1994 (Offer), the largest generators agreed to sell power plants. However, the two major generators gamed this agreement by putting anti-competitive earn-out clauses into their power plant sales terms (raising the marginal costs faced by the new owners and forcing them to bid higher). Merger  See the discussion in Newbery (2005) and Jamasb and Pollitt (2005).  See https://www.ofgem.gov.uk/gas/retail-market/retail-market-monitoring/understanding-­ profits-­large-energy-suppliers. Ofgem requires the production of consolidated segmental accounts which show the profits of the large integrated firms in each segment of the GB market. 24  See https://www.ofgem.gov.uk/gas/wholesale-market/european-market/remit 25  See https://assets.publishing.service.gov.uk/media/576bca94ed915d622c000077/appendix-­4-1-­ market-power-in-generation-fr.pdf 22 23

166 

M. G. Pollitt

investigations (MMC 1996a, b) in 1996 by the competition authority of the largest generators’ proposals to buy retailers resulted in the mergers being blocked: this led to further generation divestitures before the mergers were allowed. These asset sell-offs eventually led to wholesale power prices collapsing in early 2001.26 Meanwhile the 1997–99 investigation of pool arrangements27 led to the New Electricity Trading Arrangements (NETA) from 2001, which replaced the Pool with self-dispatch and a balancing market. The GB experience illustrates the importance of creating sufficient competition prior to the opening of the market. The creation of five equally sized fossil fuel generators (rather than two) in Great Britain in 1990 with a range of different power plants would have saved years of regulatory intervention to establish a competitive wholesale market. China has ample opportunity to learn from the UK and reorganise the ownership of its largely state owned electricity generation sector prior to full market opening. This would significantly improve its subsequent prospects for competition. As in all markets for heavily standardised products, there is a need to monitor withholding and signalling of prices. The promotion of competition via new entry and interconnection is important, as monitoring and price regulation are less satisfactory than actual competition. Regulation of some bidding behind local constraints is likely to be necessary (in the UK withholding of generation capacity in Scotland has been an issue, behind a significant transmission constraint between Scotland and England). Rapid enforcement action against anti-competitive behaviour is helpful, in order to limit the damage done by it and to increase the deterrent effect of enforcement. The stimulation of competition may mean that divestiture of generation plant by price-setting incumbents may be necessary. The independence of the SO from generation and retail is necessary in order to prevent (or highlight) anti-competitive actions by the SO aimed at promoting its own generation.

 See Evans and Green (2003) for analysis of what caused the fall in wholesale prices around this time. 27  See, for example, Offer (1998). 26

4  How Industrial Electricity Prices Are Determined… 

167

The reform of the electricity generation sector in Great Britain did raise the sort of ‘social stability’ issues that are important in China. Collusion to raise the price of generation from existing generators and divestitures by the incumbent generators probably accelerated the run-­ down of coal-fired generation in favour of gas (see Newbery and Pollitt 1997; Newbery 2005). The government attempted to slow the reduction in the use of high-priced domestic coal by privatised generators by signing preferential contracts for domestic coal for the first few years after the creation of the market. Subsequently, it announced a ‘gas moratorium’ in 1997—which made it more difficult to get planning permission for new gas-fired power plants—in an attempt to slow build of gas-fired generation, to promote coal use in electricity generation.28 Renewables support especially for onshore wind in Scotland and offshore has been promoted partly as a way of supporting Scottish development and northern ports (e.g. Hull). The withdrawal of financial support for onshore wind in England (from 2017) was also about local residents’ opposition to the siting of wind parks. Carbon pricing was initially kept low in 2013–1429 so as not to accelerate run-down of coal, but has since then has strongly favoured gas-fired generation (as we discuss below). Following the reforms in 1990, nuclear generation was unable to cover its long-run costs in the new electricity market (though its participation in the wholesale market strongly incentivised it to lower its operating costs and increase its output, which it did very effectively), but the whole retail base was subject to a levy which built up a fund to finance long-­ term nuclear liabilities. Falls in the electricity price in 2001–02 caused the financial collapse of the nuclear generator British Energy, which was successfully rescued by the government and returned to the private sector (at a profit to the government).30 Several additional aspects of the GB wholesale power market are important for China as it proceeds with wholesale power markets. All generation is in the market (including all renewables and nuclear), not  The moratorium was short-lived and not particularly effective. It was abandoned in mid-1999 and was never a complete ban. 29  See Hirst (2018) for a discussion of the history of carbon price floor in the UK. 30  See Taylor (2007, 2016). 28

168 

M. G. Pollitt

some of the generation as with many of the current round of pilot markets.31 Demand participates directly in the market and bidding is two-­ sided in the sense that generators and loads participate in wholesale markets. There are a range of contracts between retail and generation, not just one type of financial product (e.g. monthly contracts as in Guangdong up to 2019, or day-ahead contracts in California pre-crisis). There are strong incentives on generators to make least cost plant available. Plants are dispatched by the system operator on the basis of declared availability and least cost adjustments. There is complete transparency on government attempts to influence dispatch and investment (e.g. via coal contracts or bans on certain types of generation). In wholesale power markets active competition and regulatory policy have been very important in promoting competition. While spot markets are relatively easy to set up, futures markets for electricity take time and have been considered problematic.32 This is because liquidity is an issue in a market with important underlying physics which limits the ease of modelling future prices. Physical delivery has to occur in real time and stocks do not exist in electricity markets. This means that electricity markets are difficult to model years ahead and incumbent generators and retailers are strongly favoured as participants in electricity futures markets. Financial electricity futures markets do not necessarily bring major benefits to electricity consumers as they primarily exist to serve the needs of financial investors. Financial instruments should themselves be of limited interest to electricity regulators and should primarily be the concern of financial regulators. The GB market is equivalent to provincial market in China. It is important to say that GB sits within an increasingly integrated single electricity market across Europe. Beginning in 2006 France, Belgium and the Netherlands coupled their day-ahead markets, in the first cross-border coupling. This aimed to make more efficient use of available cross-border transmission capacity and ensure that the available capacity was allocated efficiently to ensure that it was fully utilised in the correct direction (i.e. flowing from high- to  See Chap. 3 on Guangdong.  See Ofgem (2016) as an example of regulatory concerns about market liquidity.

31 32

4  How Industrial Electricity Prices Are Determined… 

169

low-price areas). By 2015, Multi Regional Coupling had occurred between 19 European countries and covering 85% of the European single market’s electricity consumption. This process is known as Market Coupling and was facilitated by the European Commission.33 Spot markets across Europe are subject to market coupling via the EUPHEMIA algorithm,34 whereby prices are the same in different spot markets in the absence of transmission constraints. Extending markets over wider areas has benefits in terms of increased market efficiency35 and where there is political support for price convergence (particularly in the low-price areas where price convergence might raise wholesale prices). GB is a high-price area within Europe and has hence been keen to integrate its electricity market with that in northern Europe. However, this work of integrating national markets in Europe has been a slow process, particularly in the area of ancillary services where it remains a work in progress. It is important to point out that market coupling and the associated cross-border transmission flows are a function of the competition within national markets on which they rely. This is because physical dispatch is still carried out at the national system operator level, taking into account likely flows on interconnectors as one source of supply and/or demand. Given that cross-border transfer capacity is relatively limited for most European states, including the UK, it is changes in domestic supply and demand conditions that continue to be the major determinant of real-­ time wholesale electricity price market prices.

4.5 Retail Margins The other main component of the industrial price that is competitively determined is the retail margin. This is the mark-up on top of all of the other cost elements—wholesale prices, network charges and taxes and levies—that retailers add to cover their own costs.  See https://www.next-kraftwerke.com/knowledge/market-coupling  See Pollitt (2018a) for a discussion of the development of the single electricity market in Europe. 35  See Mansur and White (2012) on the benefits of extending the PJM market. 33

34

170 

M. G. Pollitt

So, what do retailers (suppliers) do in GB? They contract for wholesale power in spot and forward markets, hedge their physical contract position with financial contracts, meter the consumption of their customers, advertise and switch customers, provide customer services such as electrical equipment testing and monitoring and decide on and offer retail tariffs. It is important to understand that in fully liberalised power markets retailers bill for the full price of power, pay regulated transmission and distribution charges to network companies, accept and manage the non-­ payment risk of final customers, fulfil any social obligations imposed on tariffs (e.g. low use tariffs) and promote energy efficiency (e.g. via schemes like Carbon Emissions Reduction Target (CERT) and Energy Company Obligation (ECO) in GB).36 The GB retail electricity market was opened in stages via the removal of retail ‘supply’ price controls37: from 1990, 1 MW+ customers could choose supplier; from 1994, all 100 kW+ customers could choose supplier (i.e. all half-hourly metered customers); and from 1998 to 1999, all customers (i.e. non-half-hourly metered) could choose supplier. As market opening progressed significant horizontal and vertical (with generation) reintegration occurred involving retail companies, as noted above. A very significant driver of competition was the entry of the former gas monopoly, British Gas, into the electricity market. By 2002 it was the largest supplier of electricity. There has been significant market innovation since 1990: many final customers (~40% [i.e. ~10 million consumers]) have dual fuel—electricity and gas—direct debit tariffs; and there are a wide range of fixed, capped, green, social tariffs available in the market. However, the regulator has been concerned about state of supply competition (see Pollitt and Haney 2014). Ofgem launched a Competition Probe in 2008 following large price increases—which were primarily due to commodity price rises—finding no evidence of cartels but introducing new protections for vulnerable customers. These concerns resulted in the Retail Market Review in 2011, and this eventually led to the energy markets (electricity  See https://www.ofgem.gov.uk/environmental-programmes/eco/overview-previous-schemes (Accessed 27 November 2018) for more information. 37  See Henney (2011). 36

4  How Industrial Electricity Prices Are Determined… 

171

and gas) being referred to CMA—the general competition authority—in June 2014. There have been important developments with retail competition in electricity. In Europe EU directives on unbundling have been very significant in promoting retail competition and competition from gas incumbents has been important for stimulating retail competition generally in electricity markets. Small and medium-sized companies (SMEs) have been inactive in the retail market (as they often consume less electricity than a typical household) and this market segment has been subject to some limited re-regulation of their retail tariff, following CMA 2014–16 inquiry. There has been an ongoing concern in some countries about the degree of separation between retail and distribution businesses. New Zealand (in 1999) and the Netherlands (in 2006)38 have ownership unbundled distribution networks from retail. Texas and the UK have widespread voluntary ownership unbundling of networks and retail. The evidence from New Zealand seems to suggest that residential competition is significantly promoted by smart meters, which have brought down the time taken to switch (and the accuracy of switching) between retailers significantly.39 Some overall lessons from the retail competition experience in GB (and generally) can be identified. There has been an active market for larger non-domestic customers with smaller commercial customers regulated initially. There are concerns about small non-domestic (as well as domestic) customer inertia due to a lack of materiality or split incentives (between the bill payer and user) in buildings. There are worries about distorted incentives towards smaller retailers who have been exempted from certain social/energy efficiency obligations, and are hence able to undercut larger retailers unfairly by setting prices which are below the ‘competitive’ price. Retailing is increasingly about collecting revenue to pay government-imposed charges, which increases relative risks for larger, more responsible retailers, who correctly allow for non-payment risk. New retailers can target larger more creditworthy customers, but this  See Nillesen and Pollitt (2011) for a detailed analysis of the impact of ownership separation of the electricity distribution business from retail electricity in New Zealand. 39  Monthly switching has more than doubled in New Zealand since the roll-out of smart meters began. See the statistics at: https://emi.ea.govt.nz/Retail/Reports/Tagged/consumer-switching?_ si=v2. Accessed 27 November 2018. 38

172 

M. G. Pollitt

leaves incumbent retailers with customers who are less attractive to new suppliers or who are less able to pay off their existing bills to facilitate a switch between retailers, putting incumbents at a further disadvantage. There are a number of lessons for China from GB’s experience with a competitive retail electricity market. Competition from outside the electricity market was extremely significant. It was British Gas’s entry from gas into electricity that substantially shook up the small commercial and residential electricity market. Recently, the oil and gas company Shell entered the GB electricity market with its purchase of First Utility, looking to respond to the potential future electrification of transport. Stand-­ alone retailers have struggled with the risks of buying short in spot and monthly markets and selling long—most customers sign up for one-year fixed price contracts and some retailers have exited. This is not necessarily a problem (some such retailers have done well and there may be real advantages to vertical integration) but it is a challenge for such retailers. China’s new retailers (e.g. in the Guangdong market pilot) are not doing retail as in GB because they do not bill the customer for the full cost of their electricity. They are more like energy service companies advising customers on purchasing cheaper wholesale power (most payment risk remains with the local grid company that continues to bill the final customer). Ideally, existing incumbents should be able to compete in a genuine retail market. One way of doing this might be to create retail businesses at the provincial or sub-provincial level within SGCC and CSG and allow these retailers to compete within and across their current geographical limits. Retail competition limits do not have to (and should not) limit the size of the wholesale market. Although the retail market in Great Britain opened up to full competition gradually, retailers purchasing electricity on behalf of their regulated customers still participate fully in the wholesale market. Thus, China needs to find a way of getting regulated customers into the wholesale market (e.g. via procurement auctions for default contract retail customers). Future smart energy retailing business models that combine retail contracts, energy equipment sales and maintenance and energy data analytics will require sophisticated retailers (as can be observed in Great Britain) able to offer integrated solutions including metering and use of meter data.

4  How Industrial Electricity Prices Are Determined… 

173

Retail competition in Great Britain revealed a lot about customer preferences. Consumers who switched were primarily shown to be interested in price. However, customers were quick to complain and switch in the face of low quality and it became clear that the regulator had to intervene in response to a large negative deviation from average quality. Genuine retail competition helps with the discovery of the diversity of customer preferences. Thus customers can reveal preferences for their payment method (e.g. monthly, annually), the kind of tariff they want (e.g. green or brown) and their willingness to accept different tariff structures (e.g. flat tariffs, peak pricing, time of use or real-time prices). Over time, retail competition reveals what sorts of advertising methods are acceptable (e.g. door-to-door selling of residential tariffs was restricted in the UK) and social concerns about tariff fairness.

4.6 Regulated Network Charges Determination Before discussing each of the regulator-determined elements of the industrial price we need to discuss the general background to how are network charges—for both transmission and distribution—determined in GB. The total level of revenue allowed to be recovered is set by the regulator for transmission, direct system operation costs and distribution-related charges. Approved tariff methodologies then apportion this total among different customer groups to set individual prices for these services that form the charges that retailers pay on behalf of their customers. The basics of the process by which total revenue for transmission and distribution are derived are similar; we consider this first. The UK uses ex ante regulation and sets a base revenue formula and associated quality of service incentives for a fixed period in advance. This gives rise to strong incentives to perform against the formula. Ofgem, the independent energy regulator, is responsible for network charges and these are determined without direct reference to the central government. Ofgem is an Independent Regulatory Agency (IRA) with list of statutory duties. Independence involves the fixed term

174 

M. G. Pollitt

appointment of its CEO, chair and board consisting of executives and independents. Its primary functions, laid out its governing legislation, are: the promotion of competition and non-discriminatory access (as an agent of competition authority) to the grid; the regulation of the level and structure of network charges (Ofgem oversees periodic price control review process); and independence to ensure investor interests are protected and arbitrary government interference is made more costly. It is important to emphasise that Ofgem is a creature of legislation (Electricity Act, Gas Act, Competition Act) and it is, largely, independent of government. Although its board members are appointed by the  Secretary of State responsible  for  energy, the regulator answers to Parliament. It is intended to be an independent voice for economic analysis of the interests of electricity consumers. This is a key safeguard for company shareholders. For instance, if a future government wanted to renationalise some of the companies and/or sequestrate private investment, it will be the independent regulator—assuming it initially remains in place—that will likely identify the detriment to consumers of any reneging on commercial agreements. Its decisions are subject to appeal. Companies and affected third parties can appeal decisions to the Competition & Markets Authority (also largely independent of government) or seek judicial review of the process (from the independent judiciary). It is duty bound to consider the need for licensees—generators, network companies and suppliers—to fund the obligations upon them. This is not a guarantee that any company costs will be covered but a general assurance that efficient costs will be covered. For monopolies in the sector (such as the network companies), Ofgem attempts to simulate competition (with the use of rewards as well as penalties). Ofgem has a significant budget and resources: £90 m in 2017–18.40 It has a benefit-cost ratio of 87 to 1 (by its own calculation!41). It has 816 staff, of which 401 are in regulation, 273 in E-Serve, 142 in corporate functions. E-Serve administers various government programmes towards energy including energy efficiency, renewables support and social  See Ofgem (2018b), https://www.ofgem.gov.uk/publications-and-updates/ofgem-annual-reportand-accounts-2017-18 41  See Ofgem (2018c), https://www.ofgem.gov.uk/system/files/docs/2018/07/consumer_impact_ report_-­_published0307.pdf 40

4  How Industrial Electricity Prices Are Determined… 

175

programmes (including ROCs, FITs, ECO and WHD). Ofgem raises money from licenced electricity companies and administration fees and is self-­funding. It is subject to a 15% real terms reduction in its own funding by 2019–20 (set in 2015). It is staffed by well-paid civil servants. The importance of appropriately resourced regulators in successful electricity market reforms is emphasised in Pollitt and Stern (2011). Ofgem and its predecessor regulator Offer have accumulated significant experience with the regulation of network companies, where network charges have been determined in successive price control reviews since privatisation in 1990. Distribution price control reviews have (or will) reset prices from: 1995, 2000, 2005, 2010, 2015 and 2023. Transmission price control reviews have (or will) reset prices in: 1993, 1997, 2001, 2007, 2013 and 2021. Until 2010, price-cap (RPI-X) regulation in GB was explicitly designed to avoid the asset gold-plating that was observed under rate of return regulation, as used in the US. It was designed by Stephen Littlechild (the first independent electricity regulator in GB from 1989) for BT (the former monopoly fixed line telephone network operator) to facilitate a transition to a competitive unregulated market and to mimic the effect of competition. Under RPI-X the regulator collects data from the regulated utility on forecast efficient operating costs Ot; regulatory asset values, including investment plans Bt; depreciation Dt; and demand forecasts. It then determines the revenue required: Rt = Ot + rBt + Dt, where r is average cost of capital. Looking at the difference between the efficient level of revenue required and the actual revenue of the firm allows an X factor to be identified which is the scope for annual reductions in revenue. RPI-X refers to the fact that revenue is uprated by a measure of inflation (RPI in the UK) and reduced by an annual ‘productivity’ factor, X.42 The basic characteristics of OFGEM RPI-X approach in GB involved fixing the revenue required in a five-year (usually) control period for each electricity distribution company and each electricity transmission company. An initial consultation document was normally issued 18 months  This factor was intended to include all relevant factors which might drive the level of efficient revenue (including the costs of quality, or the relative movement in labour/capital costs vis-à-vis general inflation). 42

176 

M. G. Pollitt

before end of current price control period. Several subsequent documents refined the calculation of the required revenue with responses invited each time. Responses were placed in public domain unless marked confidential. A final document was issued by the regulator within six months of end of current control period. Regulated companies then have one month to appeal to competition authority (originally the Monopolies and Mergers Commission [MMC] and then the Competition Commission [CC]) if unhappy with proposals at this stage. There a number of key factors in the price control process. These include the regulatory asset base (RAB), on which the company is allowed to earn a return. Establishing an initial value for this is difficult, but subsequent updating is relatively straightforward on the basis of agreed additions to the capital base and allowed depreciation. The allowed rate of return or weighted average cost of capital (WACC) is calculated depending on the appropriate risk factor and gearing ratio. The efficient level of operating expenditure (opex), which may be subject to capital expenditure (capex) trade-off, is calculated on the basis of comparative benchmarking. And capex itself requires careful auditing as to whether the proposed investments are both necessary and being efficiently done. Figure 4.4 shows a regulated firm with starting revenue in 2010 against its efficient level of revenue and different scenarios for X factors to reduce its revenue to the efficient level by 2015. A central part of the regulation of the required expenditure of the regulated company has been the benchmarking of the actual performance. This requires a set of comparable companies, and enough data to identify important cost drivers. It is also important to predict the movement in the frontier level of performance over the upcoming price control period. The regulator in setting prices therefore needs to: identify a comparator group of firms; identify a range of efficiency measurements; identify the inputs, outputs and environmental variables to be taken into account of in the analysis; collect data on consistent basis; conduct the analysis; generate efficiency differences; generate efficient cost predictions for each firm and set X on the basis of the difference between actual and efficient costs. A difficult part of setting network charges is getting the right incentives for investment. Benchmarking has been used extensively for opex, but it is hard for capex. This is because capex is lumpy and the exact timing of when it should be done can be difficult to predict in advance.

177

4  How Industrial Electricity Prices Are Determined…  $

X factor 1

Actual Revenue 2010 X factor 2

Frontier Shift Actual Opex Efficient Opex Efficient Revenue Requirement

Depreciation WACCx RAB 2010

2015

time

Fig. 4.4  The impact of X factors on the revenue of a regulated firm

Ofgem have scrutinised  the investment plans of the companies and approved a baseline level of capex with incentives to economise on actual capital expenditure using a form of menu regulation and cost sharing.43 This involves companies that accept a lower baseline revenue getting stronger incentives to cut their costs, whereby they keep a larger share of any savings relative to the baseline.44 One mechanism that Ofgem has used is ‘menu regulation’, which is illustrated in Table 4.2. The DNO submits a baseline figure for capital expenditure over the price control period in its business plan. This figure is then audited by a firm of consulting engineers working for Ofgem. This produces a ratio DNO:Ofgem which is 100 if the expenditure submitted by the regulated firm is the same as the figure from Ofgem’s consultants. The higher the ratio, the less ambitious the firm is assessed to be in controlling its own expenditure. In return for the firm being more  For a description of Ofgem’s regulatory process for electricity distribution firms, see Jamasb and Pollitt (2007). 44  See for an example of the menu regulation scheme, Ofgem (2009a, p. 120). 43

7.19 4.81 2.44 0.06 −2.31 −4.69 −7.06 −9.44 −11.81 −14.19 −16.56 −18.94

7.5 5 2.5 0 −2.5 −5 −7.5 −10 −12.5 −15 −17.5 −20

Source: Ofgem (2009a, p. 120)

47.50% 101.25 1.84

50.00% 100 2.5

Incentive rate Allowed expenditure Additional income Actual expenditure 90 95 100 105 110 115 120 125 130 135 140 145

105

100

Ratio of forecast to baseline

Table 4.2  Menu regulation

6.75 4.5 2.25 0 −2.25 −4.5 −6.75 −9 −11.25 −13.5 −15.75 −18

45.00% 102.5 1.13

110

6.19 4.06 1.94 −0.19 −2.31 −4.44 −6.56 −8.69 −10.81 −12.94 −15.06 −17.19

42.50% 103.75 0.34

115

5.5 3.5 1.5 −0.5 −2.5 −4.5 −6.5 −8.5 −10.5 −12.5 −14.5 −16.5

40.00% 105 −0.5

120

4.69 2.81 0.94 −0.94 −2.81 −4.69 −6.56 −8.44 −10.31 −12.19 −14.06 −15.94

37.50% 106.25 −1.41

125

3.75 2 0.25 −1.5 −3.25 −5 −6.75 −8.5 −10.25 −12 −13.75 −15.5

35.00% 107.5 −2.38

130

2.69 1.06 −0.56 −2.19 −3.81 −5.44 −7.06 −8.69 −10.31 −11.94 −13.56 −15.19

32.50% 108.75 −3.41

135

1.5 0 −1.5 −3 −4.5 −6 −7.5 −9 −10.5 −12 −13.5 −15

30.00% 110 −4.5

140

178  M. G. Pollitt

4  How Industrial Electricity Prices Are Determined… 

179

ambitious the firm is offered higher baseline revenue and more generous sharing arrangements—the incentive rate—should costs be lower than the baseline amount. Looking at Table  4.2, the numbers highlighted show the impact of actually delivering on the baseline level of expenditure identified by the firm. Assume that the numbers in the body of the table are millions of pounds and that Ofgem’s assessment of the required expenditure is £100 m. A firm with plans to control costs in line with Ofgem’s assessment would get its baseline revenue of £100 m plus a bonus of £2.5 m, that is, £102.5 m. If it managed to spend less than £100 m it would share the savings at a rate of 50:50 between itself and its customers. Thus, if it actually spent £90 m, it would get an additional bonus of £5.0 m for a total bonus of £7.5 m; that is, it would be allowed to raise £97.5 m from its customers. By contrast a firm that requested baseline revenue of £140  m would only be given a starting revenue £110  m, subject to a penalty of £4.5 m, that is, £105.5 m. It would only share savings in ratio 30:70 with its customers. Therefore, if it spent £90  m, it would have saved £20 m relative to its baseline revenue and hence keep £6 m, creating at net bonus of £1.5 m; that is, it would be allowed to raise £91.5 m from its customers. By contrast if the first firm spent £140 m (against saying it would only spend £100 m) it would be subject to a penalty of £17.5 m (50% of the overspend of £40 m less the baseline bonus) and only be allowed to raise £122.5 m from its customers in revenue. If the second firm spent £140 m it would have a penalty of £13.50 m and be allowed to raise £126.50 m. In practice regulated firms have been assessed to have lower DNO:Ofgem ratios and high incentive rates (above the 50% rate); that is, this menu has incentivised firms to moderate their expenditure plans. At the same time as giving strong incentives to reduce costs there are strong incentives to improve quality of service. At the distribution level these strongly incentivise reductions in customer interruptions (CMI) and customer minutes lost (CML) per year. Table 4.3 shows the situation of the incentives for CML reduction in the current price control period. The first table shows the target number of CML for each year and each distribution company. The second table shows the symmetric reward/ penalty for deviations from the target. The third table shows the

180 

M. G. Pollitt

Table 4.3  Quality of service incentive scheme Licensee 2015/16 2016/17 2017/18 2018/19 2019/20 2020/21 2021/22 2022/23 Duration of unplanned customer interruptions term targets ENWL 40.6 39.8 39.1 38.3 37.6 36.9 36.2 35.5 NPgN 54.8 53.7 52.7 51.7 50.7 49.7 48.8 47.9 NPgY 57.5 56.3 55.2 54.1 53.0 52.0 50.9 49.9 LPN 38.8 38.1 37.5 36.8 36.2 35.6 35.0 34.4 SPN 45.5 44.5 43.5 42.6 41.6 40.7 39.8 39.0 EPN 48.0 47.0 45.9 44.9 43.9 43.0 42.1 41.2 SPD 42.2 41.3 40.5 39.7 38.9 38.1 37.4 36.7 SPMW 35.1 34.3 33.5 32.8 32.1 31.3 30.6 30.0 SSEH 53.9 52.8 51.6 50.5 49.2 47.7 46.6 45.6 SSES 48.1 47.1 46.2 45.3 44.4 43.5 42.6 41.8 Duration of customer interruptions term incentive rate (£m per CML, in 2012/13 prices) ENWL 0.89 0.89 0.89 0.89 0.89 0.89 0.89 0.89 NPgN 0.60 0.60 0.60 0.60 0.60 0.60 0.60 0.60 NPgY 0.86 0.86 0.86 0.86 0.86 0.86 0.86 0.86 LPN 0.86 0.86 0.86 0.86 0.86 0.86 0.86 0.86 SPN 0.85 0.85 0.85 0.85 0.85 0.85 0.85 0.85 EPN 1.34 1.34 1.34 1.34 1.34 1.34 1.34 1.34 SPD 0.75 0.75 0.75 0.75 0.75 0.75 0.75 0.75 SPMW 0.56 0.56 0.56 0.56 0.56 0.56 0.56 0.56 SSEH 0.28 0.28 0.28 0.28 0.28 0.28 0.28 0.28 SSES 1.12 1.12 1.12 1.12 1.12 1.12 1.12 1.12 Revenue exposure to interruptions incentive scheme term (£m, in 2012/13 prices) ENWL 22.4 22.4 22.4 22.4 22.4 22.4 22.4 22.4 NPgN 16.5 16.5 16.5 16.5 16.5 16.5 16.5 16.5 NPgY 22.4 22.4 22.4 22.4 22.4 22.4 22.4 22.4 LPN 21.8 21.8 21.8 21.8 21.8 21.8 21.8 21.8 SPN 22.4 22.4 22.4 22.4 22.4 22.4 22.4 22.4 EPN 33.5 33.5 33.5 33.5 33.5 33.5 33.5 33.5 SPD 22.4 22.4 22.4 22.4 22.4 22.4 22.4 22.4 SPMW 23.9 23.9 23.9 23.9 23.9 23.9 23.9 23.9 SSEH 14.0 14.0 14.0 14.0 14.0 14.0 14.0 14.0 SSES 30.0 30.0 30.0 30.0 30.0 30.0 30.0 30.0 Source: Ofgem (2014a, p. 15)

4  How Industrial Electricity Prices Are Determined… 

181

maximum revenue exposure (up or down) per company. This is set at 1.8% of total revenue for CML (and 1.2% of total revenue for CMI). It is important to reiterate that Ofgem has employed ex ante (incentive) regulation, where revenue formula are set in advance and companies have an incentive to deliver services efficiently (at low cost) and at high quality (because various quality measures—such as customer minutes lost—are also subject to baseline quality targets which if exceeded allow the company to increase its revenue).45 This is the best simulation of competition. There are strong incentives to outperform ex ante revenue allowances. Companies can improve returns to shareholders within each regulatory period.46 This also reveals information for regulators to better set allowances and pass efficiencies to consumers in the next regulatory period. This system removes regulatory uncertainties and overheads inherent in ex post regulation (and the risks of regulatory micromanagement). It gives scope for innovation in opex, capex and financing costs together with internalised outputs. However the revenue formulae are tricky to set and future uncertainties remain (especially with respect to climate change and climate policy) and there is a large information asymmetry between the private knowledge of the regulated companies and the regulator. As we will document later the RPI-X regulation of transmission and distribution charges was very successful in GB. However the system was changed in 2010. The background to this change was changing circumstances (Pollitt 2008), including: rising investment needs in electricity distribution (+48%, 2005–10 vs. 2000–05) and in electricity transmission (+79%, 2007–12 vs the previous period); network tariff charges being increasingly driven by capex not opex and network capex being increased by rising amounts of subsidised renewables connected to the system. A review was announced by Ofgem in 2008—the RPI-X@20 review (see Ofgem 2009c)—focussing on customer engagement, sustainability and the scale and scope of innovation. The result of this review was a new system of regulation called RIIO: Revenue = Incentives + Innovation +  See Ofgem (2009b, p. 63) for a discussion of quality incentives.  See Ofgem (2014b, p. 44), which shows that electricity distribution companies could potentially double their real returns on assets if they received the maximum possible incentive payments. 45 46

182 

M. G. Pollitt

Outputs. Under RIIO there was more emphasis on incentives, promoting network innovation and on a wider range of outputs (such as stakeholder satisfaction). Notable changes included more money for innovation, a longer price control period (eight rather than five years) and a greater emphasis on total expenditure (totex) not just capex and opex. However, RIIO is more of an evolution of RPI-X than a revolution in the way network charges are determined. One interesting observation for China is that Great Britain does have some cross-subsidies between areas in transmission and distribution charging. These always exist within a single service area (e.g. within UK distribution network operator [DNO] areas and the regions of Guangdong). Transmission and distribution charges are not fully cost reflective in this sense. However, in the UK they are not aimed at promoting economic development in underdeveloped regions as in China. It is a good idea not to distort the market elements of the electricity price to deliver a locational cross-subsidy. It is better to use transmission and distribution charges to do this. This is easy to administer and pass on to customers; the difficulty is that it may distort connection location decisions. Another way to deliver lower electricity prices to particular regions is to simply levy a per MWh charge on everyone and reduce final price of electricity for customers in the favoured region; however, it is more difficult to ensure pass-through. At the provincial level in China, one or other of the trading provinces could tax imports/exports of cheap electricity to finance the cross-subsidies. For example, Guangdong could tax hydro imports from Yunnan, or Yunnan could tax exports to Guangdong. These taxes would raise revenue without distorting the wholesale price, and the revenue could be used to subsidise electricity sales to favoured areas.

4.7 Transmission Charges RPI-X regulation was very successful in reducing transmission charges. These fell by around 40% between 1990 and 2005 as a result of gently rising demand, falling operating costs and modest capital expenditure. Figure 4.5 shows the reduction in real operating costs at National Grid Electricity Transmission (NGET), the largest transmission company in

183

4  How Industrial Electricity Prices Are Determined…  £m 2007/08 Prices 700 600 500 400 300 200

17/18

16/17

15/16

14/15

13/14

12/13

11/12

10/11

09/10

08/09

07/08

06/07

05/06

04/05

03/04

02/03

01/02

00/01

99/00

98/99

97/98

96/97

95/96

94/95

93/94

92/93

91/92

0

90/91

100

Fig. 4.5  NGET controllable operating costs. (Source: National Grid)

£m 2007/8 prices 1,200 1,000

CEGB Transmission

NGET

800 600 400

0

79/80 80/81 81/82 82/83 83/84 84/85 85/86 86/87 87/88 88/89 89/90 90/91 91/92 92/93 93/94 94/95 95/96 96/97 97/98 98/99 99/00 00/01 01/02 02/03 03/04 04/05 05/06 06/07 07/08 08/09 09/10 10/11 11/12 12/13 13/14 14/15 15/16 16/17 17/18

200

Fig. 4.6  Transmission investment. (Source: National Grid. Pre-1991 figures are based on CEGB information sources and are adjusted to reflect one-off investment in the GB-France interconnector)

GB (accounting for around 80% of total GB transmission revenue). Figure  4.6 shows the evolution of real investment at NGET over the same period, which has been financed successfully in the privatised company. Meanwhile the transmission system has remained very reliable, with no increase in loss of supply incidents (Fig. 4.7) or in energy not supplied (Fig. 4.8). Note that total supply is of the order of 300 TWh, so the level of energy not supplied is trivial. Overall transmission company revenues are determined by the regulator as discussed in the previous section. The charges which customers of the transmission system pay are paid via: connection charges, which are

184 

2

6

5

4

3

2

4

3

08/09

09/10

10/11

2

7

3 2

5

3 2 17/18

8

4 1

16/17

9

1 3

15/16

7

2

7

14/15

2 1

07/08

5

06/07

99/00

1

5

05/06

98/99

6

04/05

97/98

10

03/04

4

8

02/03

9

5

01/02

10

2

00/01

1

1 3

96/97

5

4 95/96

6

2

3

13/14

4

4

94/95

2

5

12/13

4

Custom connections

11/12

Secure connections (incentivised)

93/94

No. 16 14 12 10 8 6 4 2 0

M. G. Pollitt

Fig. 4.7  Loss of supply incidents at NGET.  Note: Custom connections are for certain industrial customers who accept higher levels of interruption in return for lower tariffs. (Source: National Grid)

17/18

16/17

15/16

14/15

13/14

12/13

11/12

10/11

09/10

08/09

Custom connections

07/08

06/07

05/06

04/05

03/04

02/03

01/02

00/01

98/99

97/98

96/97

95/96

94/95

93/94

1600 1400 1200 1000 800 600 400 200 0

99/00

Secure connections (incentivised)

MWh

Fig. 4.8  Energy not supplied by NGET. (Source: National Grid)

charged to generators and loads/distributors; transmission use of system charges (TNUoS), which are charged to generators and loads (generators pay per MW and loads pay per MW and per MWh); and international interconnector charges, which are largely paid for by users via arbitrage revenue. Transmission losses are recovered via a transmission loss multiplier which adjusts metered volumes (allocated 45% to generators, 55% to retailers), calculated by Elexon. There are incentive payments for quality of service. We discuss these elements in more detail below. Connection charging is an asset-based charge levied on users to recover the costs, with a reasonable rate of return, of providing assets for connection to the GB transmission system. The charges relate to the cost of assets installed solely for and only capable of use by an individual user. They are charged by asset, taking account of asset value, asset age, site-specific maintenance and the costs of running the transmission system.

4  How Industrial Electricity Prices Are Determined… 

185

Generation TNUoS charges reflect the incremental cost of facilitating generation on the transmission network: the higher the  network requirement the higher the charge; this incentivises efficient location. It is charged to all Directly Connected Generation; interconnectors do not pay; but the zonal element is also paid by Non-Licence Exemptible Embedded Generation. Annual Chargeable Capacity is based on the maximumtransmission entry capacity (TEC); in zones where the price is negative, the output is taken to be the average of three ‘proving runs’. Demand TNUoS charges reflect the incremental cost of facilitating offtake from the transmission network: the higher the network requirement, the higher the charge. This incentivises efficient location of demand and/or offsetting distributed generation. It is charged to all offtakes, but international interconnectors do not pay. Annual Chargeable Capacity is based on half-hourly metered consumption. This is assessed on the ‘Triad’ which equals the average of consumption in the three half-hours of largest system demand between November and March separated by ten days. Half-hourly metered consumers can reduce offtake charges by managing demand in ~20 likely peak demand periods per winter. Non-half-hourly metered consumption is charged based on energy taken between 16:00 and 19:00 throughout the year. The current split between generation and demand charges is shown in Fig. 4.9.

Fig. 4.9  TNUoS 2018/19 regulated revenue. (Source: National Grid)

186 

M. G. Pollitt

The calculation of the price signal in the zonal charging for transmission capacity is built up as follows: take a base network; adjust for the winter peak, contracted generation and forecast demand; measure the flow on each line and the total MW (Tm); add a MW on at each node and observe the new flow MW (Tmn); calculate nodal cost of accommodating the increase, Tm – TMn = Ti (MWkm); applying an historic cost of providing a MWKm, £/MWkm expansion constant (EC); use this to calculate TNUoS and cost per MW, Ti × EC = £/MW; and group into zones for actual charging. The 27 generation TNUoS zones are recorded in National Grid (2018, p. 47) and the 14 demand TNUoS zones are recorded in National Grid (2018, p.  48). The current tariff rates vary considerably by zone. In 2018–19, it cost a generator £20.89/kW to be connected in the north of Scotland, but it was paid £11.26/kW if it was connected in Greater London.47 Similarly, a load connected in Greater London paid £54.91/kW, against only £26.30/kW in the north of Scotland.48 The reconciliation between the allowed total revenue (in the previous section) with the revenue actually raised is as follows. The annual tariff is determined in January before relevant charging year starting April. It seeks recovery of allowed revenues (and best endeavours not to exceed allowed revenues). It is based on a forecast of the charge-base (TEC, half-­ hourly metered triad demands and non-half-hourly metered demand). Users are initially charged monthly on the basis of the forecast charge-­ base. These revenues are subject to reconciliation with the actual demands/ consumption as the energy market settlement is finalised. The tariff values are not adjusted. Total revenue from reconciled charges may be larger or smaller than the allowed revenues. The system operator must reduce subsequent allowed revenues (two years after the charge year) with interest on any over-recovery. Penal interest is payable if actual recovery exceeds 2.75%. The system operator may recover under recoveries (by increasing recovery two years after charge year). Interest costs are not recoverable if revenue recoveries are less than 94.5%. If large under- or

 See National Grid (2018, p. 13).  See National Grid (2018, p. 8).

47 48

4  How Industrial Electricity Prices Are Determined… 

187

Revenue Adjustment (%)

1.5% 1.0% 0.5% 0.0% -0.5%

0

100

200

300

400

500

600

700

800

900

-1.0% -1.5%

Loss of Supply Volume (MWh)

Fig. 4.10  NGET’s 2008/09 reliability incentive scheme. (Source: National Grid)

over-recoveries occur in successive years, the SO must explain the reasons for this and seek the regulator’s permission for corrective actions. The transmission revenue is subject to a Transmission Network Reliability Incentive. This was introduced following high-profile interruptions in London and Birmingham in 2003. This is an opportunity to earn up to 1% additional revenue (currently, around £11–12  m) for annual loss of supply below the annual average. It includes the potential to lose up to 1.5% revenue (£17–18 m) for annual loss of supply above the annual average. The nature of the incentive is illustrated in Fig. 4.10. International interconnectors are regulated differently and their regulation has evolved somewhat over time. In 2001, the French interconnector (IFA) was separately licensed (from the onshore transmission system) requiring: non-discriminatory regulated third-party access (RTPA) with ‘Use it or lose it’ access rights; and compliance with European Union (EU) ‘use of congestion revenues’ requirements which means capacity sales revenues not required for ‘guaranteeing availability’ must be returned to national TSO charge-payers. In recent years, Ofgem have decided that IFA revenues are exceeding that needed to guarantee availability and have imposed a cap and revenues sharing mechanism. In the mid-2000s, BritNed (linking England and the Netherlands) was conceived of as a merchant interconnector project through a National Grid and TenneT joint venture. This cable is therefore owned by the incumbent onshore transmission companies at either end, but separate from their regular transmission system operator (TSO) activities. It finances its capital and operational costs from implicit and explicit capacity sales. As part of securing an exemption from the EU use of congestion revenues requirement, the European Commission required a cap on the returns achieved

188 

M. G. Pollitt

(measured over 25  years of planned operation). National Grid and TenneT agreed to continue the project despite the asymmetric return prospects (a cap on the maximum revenue, but no guaranteed minimum revenue). Other interconnectors have been subject to different arrangements. The Northern Ireland to Southwest Scotland (Moyle) transmission link was built and operated by a company funded by Northern Ireland transmission charge-payers (capacity sales reduce charge-payer costs). The Ireland-England (East-West) interconnector was built by the Irish Electricity Supply Board as a TSO regulated asset funded by Irish transmission charge-payers. For new links to Belgium and Norway, National Grid and partner TSOs worked with regulators to make a hybrid regulated/merchant arrangement such that there is a broadly symmetrical cap and floor on sales revenues. Other companies are considering interconnectors (e.g. ElecLink plan to use the Channel Tunnel). They have indicated they will use an asymmetric capped merchant model. We can make some observations on locational pricing and cost-­ reflectivity of transmission charges in GB.49 GB has not so far adopted locational marginal prices (LMPs) to reflect short-run transmission network constraints.50 In-line with the initial primary role of National Grid as a ‘wire provider’ in England and Wales, the initial charging methodology allocated allowed revenues to the degree users made transits across key boundaries. A review of transmission charging in 1992 identified merits in signalling short-run marginal costs but decided to signal longrun (investment cost-based) marginal costs on the basis of consistency with National Grid’s role as a ‘wire provider’ and practical issues. These practical issues included the following. The fact that in 1992 the extraction of short-run shadow prices from the complex scheduling and dispatch Pool software—in use at the time—was non-trivial. Peak network power flow patterns were relatively stable and this facilitated long-­ run use and network need predictions. Market parties wanted a transparent and stable tariff for at least the next year. Improving long-run  For a history of the discussion of locational signals within the GB power system, see CMA (2016b). 50  For discussions of locational marginal prices, see Bohn et al. (1984) and Hogan (1992). 49

4  How Industrial Electricity Prices Are Determined… 

189

signals was judged adequate for informing the location of CCGT new entrants. The incremental benefits of short-run signals were initially calculated to be moderate in a centrally dispatched market. Distribution charging follows a broadly similar long-run (investment cost-based) approach (but considerably different in detail). The introduction of generation and storage self-dispatch (with NETA), the development of variable wind, higher market-driven interconnector flows, an active demand side and increasing internal congestion in the network—due to increasing renewables—mean the case for short-run signals is increasing and a further review is imminent. Some important lessons for China from the GB experience with transmission charging include the following. Nodal prices arising from LMP algorithms are really about constraints in the transmission system. They reflect transmission constraints rather than being about the creation of a wide area electricity market. Thus nodal price differentials are primarily about sending signals around the use of the transmission system. Thus they are not about the electricity supply and demand balance across the market area as a whole. This raises the question about whether LMPs are a good way of incentivising best use and development of the network. In the UK and Europe we have mostly concluded that they are of limited value in addressing the optimisation of the network in the long run as their long-run efficiency depends on how the network company responds to the price signals nodal prices give rise to. There is some value in charging some transmission costs to generators to focus incentives on generators, rather than indirectly via loads. Locational signals can be delivered to incentivise the location of generators and loads via zonal charges or via locational marginal prices (LMPs). LMPs are volatile and may not be as good long-term signals as zonal transmission charges. Though the existence of the ability to trade financial transmission rights (FTRs) does mitigate some of the financial risk associated with LMPs, history, even in LMP jurisdictions, suggests zonal charges should be implemented first as a stepping stone to LMPs.51 LMPs do not solve the residual transmission pricing problem. There is still a 51

 See Pollitt (2012b) on the history of ISOs in the US.

190 

M. G. Pollitt

need to recover most of the fixed cost of the transmission system via another charging mechanism.

4.8 System Balancing Charges Within the overall industrial price of electricity, the charges for system operation in GB cover all of the costs incurred by the system operator. These consist of internal (staff and IT) costs and allowed profit (very small, c.£160 m per annum) and the external (procurement) costs which could be as high as £1000 m. Internal costs are subject to price cap regulation, similar to transmission. External costs are subject to market testing and incentives for their overall minimisation. In GB, both are recovered by a balancing service use of system charge (BSUoS) from generation and demand (50:50) per MWh, less imbalance charges (described below) recovered from parties.52 See Table 4.4 for the breakdown of costs and the summary of the calculation of per MWh charge to each side of the market for 2017/18. The accepted bids and offers for energy balancing are published.53 The largest component of external system operation costs arises from the balancing mechanism. Figure 4.11 summarises the place of the balancing mechanism (BM) within the pricing of energy following the end of the Pool in 2001 and the introduction of the New Electricity Trading Arrangements, and the subsequent British Electricity Trading Arrangements from 2005. Imbalance, and hence exposure to the balancing mechanism, has to be measured. Market parties register bilateral contract volumes to market and settlement system operator (Elexon) at Gate-closure (t−1  hour). Notifications update relevant (market-wide) production and consumption accounts. Physical meters are registered to particular balancing mechanism units (BMUs) who include large generators (>50 MW) and must have individual BMUs linked to production accounts. Suppliers  See https://www.nationalgrid.com/uk/electricity/charging-and-methodology/balancing-servicesuse-system-bsuos-charges 53  See https://www.bmreports.com/bmrs/?q=balancing/detailprices 52

4  How Industrial Electricity Prices Are Determined… 

191

Table 4.4  System operator services

Projection of Scheme Outturn Cost (£m) Energy Imbalance Operating Reserve BM Startup STOR Constraints - E+W, Cheviot, Scotland Footroom Fast Reserve Response Reactive Black Start Minor Components ROCOF (E&W) Black Start (non-incentivised) TOTAL BSUoS Estimated BSUoS Vol (TWh) Forecast NGET Profit/(Loss) Estimated Internal BSUoS (£m) Estimated BSUoS Charge (£/MWh)

Total 17/18 -22 92.2 1.2 88.2 374.3 10.8 97.3 139.8 78.5 0 22.5 59.2 57.7 999.7 503.1 10 164.4 2.33

Source: National Grid, MBSS_DATA_MAR18

have a BMU per distribution network all aggregated to a market-wide consumption account. Half-hour meters are compulsory for generators and large loads (>100  kW). Half-hour meters measure the flow from transmission to distribution networks. Initial settlement involves metering aggregators summing half-hourly meter values to supplier BMUs and allocating the remaining transmission to distribution system (Tx→Dx) flow to supplier BMUs in-line with supplier customer estimates. Under final settlement: non-half-hourly meter readings are allocated to half-hours using one of a number of standard profiles for specific customer classes. Residual errors between Tx→Dx flow and supplier BMUs (which include distribution system

192 

M. G. Pollitt

Forward prices

Spot prices (marked index prices)

Firm bilateral contracts: Over The Counter Forwards (90%) or Power Exchange Future (5%) Prices informed by fundamentals and risk of cashout penalties Self-dispatch

Real-time (cashout imbalance prices)

BM firm adjustments contracts for imbalance and constraints: INCs = T-1 offers (payment by SO) DECs = T-1 bids (usually payment to SO) Net cost socialised via BSUoS

Gate closure: Contracts volumes notified to Elexon Final Physical Position notified to SO

Y-1

T-1 hr

T-0

Fig. 4.11  The place of the balancing mechanism relative to real time. (Source: National Grid)

losses) are allocated pro-rata. Initial ‘cashout’ settlement (on the difference between initial meter volumes and contract volumes) is undertaken at t+28 days. Final cashout reconciliation (using final meter allocations) is at t+14 months. The system operator’s role in balancing the system is as follows. Notifications of intended physical positions (generation self-dispatch and demand forecasts) to the system operator are separate from contract volume notifications. Initial position notifications are submitted at t–24  hours. These are updated as new information emerges until the t−1 hour Final Physical Notification (FPN). Notifications are location specific to the balancing mechanism units (BMUs). The system operator will make national demand forecasts which suppliers may use to make their individual notifications. Market parties may also post balancing mechanism offers (to increase power to system) and bids (to reduce power to system) specifying the BMU of delivery—usually in pairs. The system operator acceptance of a bid or offer is a firm contract (generally no cancellation). Unwinding of a BM contract is by expiry of instruction or

4  How Industrial Electricity Prices Are Determined… 

193

Fig. 4.12  Balancing market information flows. (Source: Information from National Grid)

acceptance of a reverse trade. The system operator has discretion to trade energy with the wider energy market for physical balancing purposes (the SO is prohibited from making any speculative financial trades) and to procure specialist balancing services (ancillary services) using various platforms. The net energy position of system operator is the market net imbalance volume (NIV). Figure 4.12 shows the flow of information in balancing and settlement. Figure 4.13 shows the bids and offers for one BMU that is offering to go up and bidding to go down in 50 MW increments, at given deviations from its FPN position—which is itself not static. Initially it is offering to go up 50 MW for a payment of £27/MWh (and up another 50 MW for £33/MWh, and also offering a final 50 MW up at £140/MWh) or to go down by 50 MW for £11/MWh (i.e. it will pay the SO for the ability to reduce fuel burn while still meeting other contracts) and down another 50 MW for −£75/MWh; that is, it will require payment by the SO to cover additional costs). All the bids offered in the BM must lie within the maximum export limit (or generator and transmission equipment might be damaged) and the minimum stable export limit (so that minimum stable production levels are exceeded). Figure  4.13 shows that the SO actually required this unit to go up 100 MW relative to its FPN and hold before coming back down 50 MW to its FPN. This would have required payments in the BM of £27/MWh for the blue area of extra generation and £33/MWh for the red area of extra generation.

194 

M. G. Pollitt Max Export OFFER: £140/MWh Limit BID: £75/MWh

MW +100

OFFER: £33/MWh BID: £23/MWh

+50

OFFER: £27/MWh BID: £11/MWh OFFER: £23/MWh BID: -£75/MWh

FPN -50 -100

Half-hour trading period

Stable Export Limit

Fig. 4.13  Bids in the balancing mechanism. (Source: National Grid)

Imbalance cashout prices are calculated as follows. The system operator will enter many contracts in each half-hour trading period to balance the system, continuously matching production and consumption, establishing reserves and frequency control capabilities of required responsiveness and resolving network congestion. Some contracts will be specifically flagged by the system operator as being solely associated with system issues rather than residual energy balancing and these will be excluded from the determination of imbalance prices. The remaining buy and sell actions will be ranked in price order. Energy balancing actions are defined as the cheapest actions in the net imbalance volume (NIV) direction. The imbalance price is then determined by the average of the most expensive bids in the Price Average Reference Volume (PAR) in the NIV direction. Currently the PAR is set at 1 MWh. An example of ranked up and down bids in the balancing mechanism is shown in Fig. 4.14.54  These offers and bids are all of the actions taken by the system operator in a half-hour period. The flagged actions in blue reflect actions taken for system specific reasons which could be due to frequency control or constraint management. The unflagged actions in orange may have been wholly or partly necessary for energy balancing. The NIV is measured looking at the net impact of all actions taken by the SO. It shows that there were 475.5 MWh of offers and 245 MWh of bids taken, so the market was short by 230.5 MWh. The system imbalance price is the marginal net imbalance offer required to solicit 230.5  MWh of net supply, which in this case is £60/MWh. 54

4  How Industrial Electricity Prices Are Determined… 

195

Buy Ranked Set Key

Offer 60MWh at £320/MWh BSAA - Buy 55MWh at £170/MWh Offer 50MWh at £140/MWh

First Stage Flagged

Offer 25MWh at £120/MWh

First Stage Unflagged

BSAA - Buy 35MWh at £70/MWh Offer 0.5MWh at £70/MWh Offer 30MWh at £60/MWh Offer 70MWh at £50/MWh

Ranking order

Offer 120MWh at £40/MWh Offer 30MWh at £30/MWh Bid 30MWh at £35/MWh Bid 50MWh at £27/MWh Bid 75MWh at £25/MWh

Sell Ranked Set

Ranking order

BSAA-Sell 50MWh at £24/MWh Bid 40MWh at £23/MWh

BSAA = Balance Service Adjustment Action Fig. 4.14  Ranked bids in the balancing mechanism. BSAA = Balance Service Adjustment Action. (Source: National Grid)

System Operator Incentive Schemes ensure that the system operator minimises the external costs it imposes on the system. This applies to Given that all those parties who were out of balance are subject to this £60/MWh charge and the system operator is on average paying less than this for taking energy balancing actions, imbalance charges result in net receipts from the metered parties (hence the negative contribution to overall balancing charges in Table 4.4).

196 

M. G. Pollitt

E&W Real (07/08 £m)

Cheviot

15/16

16/17

14/15

13/14

12/13

11/12

10/11

09/10

08/09

07/08

06/07

05/06

04/05

03/04

02/03

01/02

00/01

BETTA

99/00

98/99

97/98

96/97

95/96

94/95

93/94

92/93

Incentivised

91/92

400.0 350.0 300.0 250.0 200.0 150.0 100.0 50.0 0.0

90/91

£m (07/08 prices)

intermittent renewable generators who are—correctly—exposed to their greater likelihood of being imbalanced and hence can expect to receive less average revenue per MWh as a result.55 Initially after privatisation the external (balancing) costs of operating the transmission system were passed through to suppliers and onto consumers without any market actor taking responsibility. As a result, there was a sharp rise in such costs. These costs included: congestion costs, reserve and frequency response, losses and reactive power. Following prompting by the regulator, National Grid bilaterally negotiated an external cost management incentive scheme. The resets of this scheme were subsequently overseen by the regulator. This has resulted in a series of incentive schemes by NGET to reduce operational costs. There was a sliding scale that shares costs and benefits with customers, usually over a short (one- or two-year) duration. The sharing factors + caps/collars limit risk of negative externalities on consumers and provide some internalisation of consequences of NGET investment and network asset management decisions. Figure 4.15 shows the evolution of system constraint payments, which initially fell under incentive regulation in England and Wales. These then increased when National Grid’s role was extended to cover the whole of Great Britain. Figure 4.16 shows the evolution of National Grid’s entire external costs of system operation over the period.

Scotland

Fig. 4.15  National Grid transmission system operator constraint payments. (Source: National Grid)

 See Newbery (2012) for a discussion of this.

55

197

4  How Industrial Electricity Prices Are Determined…  1000 900

Central pool (no SO incentives)

£m (2007/8 prices)

Target Outturn

Target Outturn

818 840

800

NETA (Self dispatch)

Central pool & SO incentives

BETTA (Includes Scotland)

870 Target Outturn

796 803

709

700

686

600 577

500

581

537

555 495

400 428

464 455

431

340

300

438 445

366 302

520

406

440 515445 461 450 426

260 247

200

780704 721 723 687 698 613

566

541 475 460

424

778

496

403

257

100 0

0

0

0

0

90 /9 1 91 /9 2 92 /9 3 93 /9 4 94 /9 5 95 /9 6 96 /9 7 97 /9 8 98 /9 9 99 /0 0 00 /0 1 01 /0 2 02 /0 3 03 /0 4 04 /0 5 05 /0 6 06 /0 7 07 /0 8 08 /0 9 09 /1 0 10 /1 1 11 /1 2 12 /1 3 13 /1 4 14 /1 5 15 /1 6

0

Fig. 4.16  National Grid electricity system operation external costs. (Source: National Grid)

The role of the system operator, National Grid Electricity Transmission System Operator (NGET SO), has developed over time. Pre-privatisation (i.e. before 1989) it involved real-time operation of CEGB (in England and Wales) generation and transmission assets. From 1990 to 1994 the England and Wales transmission system operator provided a central dispatch agency service to the market. From 1994 to 2000 it was a transmission system operator and central despatch agent with exposure to balancing costs. From 2001 to 2004 the New Electricity Trading Arrangements were in operation (with self-dispatch market) and the balancing incentive remained. From 2005 the British Electricity Trading & Transmission Arrangements have been in place (Scotland joins NETA). This, additionally, made National Grid responsible for system operation and balancing in Scotland. In 2014 NGET SO was appointed Electricity Market Reform delivery agent (which includes the administration of the central government Capacity Mechanism and low carbon Contracts for Differences). In 2015 NGET SO was given enhanced system planning responsibilities following Ofgem’s Integrated Transmission Planning & Regulation Project.

198 

M. G. Pollitt

In April 2019 National Grid placed the system operator in a wholly separate company, National Grid Electricity System Operator. Anaya and Pollitt (2017) have made a number of recommendations to Ofgem about how this more independent system operator might be regulated, drawing on the experiences of independent system operators (ISOs) in the US, South America and Australia. Good regulation involves not only assessing the efficient amount of revenue that the ISO requires but also ensuring the efficiency of its procurement methods (market-based) and system optimisation (procurement levels). Stakeholders (generators, network companies, retailers and customer groups) play a key role in the proposal and design of detailed implementation rules for new initiatives for the best ISOs. Sophisticated voting rules are observed and are worthy of study for the lessons they might have for GB. A high level of internal and external oversight of ISO decision making is observed which is becoming more complex and subject to high levels of uncertainty. In electricity, US ISO State of Market Reports provide excellent examples of regular updates on key recommendations for future market design. A recent issue that has arisen in system operation in Great Britain is the efficiency of the procurement of ancillary services by the system operator. Typically, the GB SO has 3–5 GW of reserve contracted a day ahead. For example, it procures a large number of ancillary services under different procurement mechanisms. It uses auctions (pay-as-bid) for balancing market (BM), firm fast response, short-term operating reserve (STOR), STOR Runway, enhanced optimal STOR, firm frequency response (FFR) (primary, secondary and high) and enhanced frequency response (EFR). It uses bilateral tenders for balancing mechanism start-up, demand turn-up, mandatory frequency response, frequency control by demand side management (DSM), FFR bridging contracts, transmission constraint management, contingency balancing reserve, max generation, intertrips, black start and SO to SO transactions. In addition, some services such as reactive power are procured at fixed prices. As these lists indicate there are a large number of ancillary service products. In 2016 there were 30 ancillary service products in GB, now reduced to around 22 (in 2019). However questions have been raised as to whether this could be reduced further (perhaps to just four: reserve, security, frequency and voltage support). Greve et al. (2018) discuss how

4  How Industrial Electricity Prices Are Determined… 

199

£ / MW/h Dynamic FFR 16 14 12 10 8 6 4 2

18 Ap r-1 8 Ju n18 Au g18 Oc t-1 8 De c18 Fe b19 Ap r-1 9 Ju n19 Au g19 Oc t-1 9

b-

17

Fe

7

cDe

t-1 Oc

Au g

-1

7

0

Fig. 4.17  Prices for firm frequency response. (Source: GB FFR Market Summary Reports from Aurora Energy Research https://www.auroraer.com)

the ancillary services (A/S) product definition needs to be clarified with too many ill-defined products. The SO needs to justify procurement quantities and express trade-offs transparently. Opportunities for gaming the system may well exist and be increasing as the products become more important, especially if there are a lack of penalties for creating A/S demand. Optimal contracts are not currently clear because of the uncertain nature of the counter-party to the SO. Distribution system operator (DSO)-TSO conflicts need to be resolved as DSOs increase their relative ability to supply A/S. Demand for ancillary services in GB has, apparently, not risen much even though RES share has risen significantly. Meanwhile prices for some ancillary services have fallen recently due to increased competition, including from electrical energy storage (EES) and from interconnectors under conditions of low demand growth (see National Grid ESO 2019). As an example of recent prices for ancillary services, see Fig. 4.17. Recently the government has introduced a capacity mechanism in Great Britain, which is administered by the system operator. Capacity mechanisms have a history in GB.  Up until 1989 there was centrally planned capacity to meet nine winters per century loss of load probability within the CEGB.  From 1990 to 2000 retail companies had a

200 

M. G. Pollitt

nine-­winters-­per-century obligation which could be discharged by purchasing energy in the Pool market. The Pool purchase price = SMP + (VoLL – SMP)∗LOLP, where SMP was the system marginal price (the market clearing price), VoLL was the value of lost load and LOLP was loss of load probability. Capacity was paid VoLL ∗ LOLP. However, there were concerns that SMP and LOLP are subject to manipulation by generators with market power. In 2001–04 under the New Electricity Trading Arrangements (NETA), there were only firm bilateral energy-only contracts and LOLP obligations removed. This has continued under BETTA. In 2012 Government Energy Market Reform (EMR) identified the need for capacity mechanism. In 2013 Regulator made a final decision on cashout pricing (incorporating a VoLL). From 2013 on, the Supplemental Balancing Reserve (SBR) and Demand Side Balancing Reserve (DSBR) were introduced to provide extra winter capacity. These schemes paid for additional emergency capacity. In 2014 the capacity market began with a t−4  years-ahead auction (for winter 2018/19) which cleared at £19.4/kW.56 However the most recent auction in January 2020 a t−3 auction cleared at £6.44/kW. The developing experience of GB offers a number of lessons for China. The system operator function is important in that it lies at the heart of the system. The SO needs an incentive to manage its own internal costs. In GB, the SO has internal revenue of £140  m p.a. subject to a 50% incentive rate. It is even more important to incentivise the SO to procure external services efficiently. In GB, the external costs are c.£850 m p.a. now subject to +/−£30 m stakeholder panel-determined incentive.57 The SO does not need to be integrated with transmission operator (TO) to function effectively. SO functions are increasingly subject to competition and market testing. The work of co-optimising wholesale energy and ancillary services markets58 (and indeed further co-optimising across wholesale power markets and network investments) remains an important work in progress in all advanced electricity systems.  There are t−1 years capacity auctions as well.  See Ofgem (2018d), https://www.ofgem.gov.uk/system/files/docs/2018/02/policy_decision_on_ electricity_system_operator_regulatory_and_incentives_framework_from_april_2018.pdf 58  See Anaya and Pollitt (2018) for a discussion of co-optimisation. 56 57

4  How Industrial Electricity Prices Are Determined… 

201

4.9 Distribution Charges Incentive regulation of distribution charges has led to a very strong relative reduction in distribution charges. Between 1995, when charges were first reset by the regulator, and 2005, prices fell by around 50% for the average distribution company in England and Wales and by more in some areas (e.g. SWEB in the southwest of England).59 Figure 4.18 shows the fall in real distribution charges—the reduction being larger than in transmission charges in England and Wales. At the same time quality of service improved substantially, with average customer minutes lost falling from over 100 minutes per year in 1990 to around 30 minutes today, as shown in Fig. 4.19. It is important to stress that overall revenues for distribution companies are determined by the regulator as discussed in Sect. 4.6. They are then charged out to individual customer groups in each distribution company area via a common charging methodology. Connection charges are charged to generators and loads/distributors according to which connection is requested and to cover sole use assets.

Index (1989/90 = 100)

120 110 100 90 80 70 60 50

89

/9 90 0 /9 91 1 /9 92 2 /9 93 3 /9 94 4 /9 95 5 /9 96 6 /9 97 7 /9 98 8 /9 99 9 /0 00 0 /0 01 1 /0 02 2 /0 03 3 /0 04 4 /0 05 5 /0 06 6 /0 07 7 /0 08 8 /0 09 9 /1 10 0 /1 11 1 /1 12 2 /1 13 3 /1 14 4 /1 5

40

Fig. 4.18  The development of real distribution revenue since privatisation average DNO in England and Wales. (Source: National Grid)

 Domah and Pollitt (2001) find significant gains for society following the privatisation and incentive regulation of the regional electricity companies (RECs) that owned the distribution and incumbent retail assets. 59

202 

M. G. Pollitt

CIs/CMLs per 100 Customers

120

Customer Minutes Lost (industry average)

100

Customer interruptions (industry average)

80 60 40 20

2016/17

2015/16

2014/15

2013/14

2012/13

2011/12

2010/11

2009/10

2008/09

2007/08

2006/07

2005/06

2004/05

2003/04

2002/03

2001/02

2000/01

1999/00

1998/99

1997/98

1996/97

1995/96

1994/95

1993/94

1992/93

1991/92

0

Fig. 4.19  Improvement in quality of service since 1990. (Source: Ofgem)

Generators need to contribute to the cost of upgrading the distribution system up to the next substation at the voltage level at which they are connected (so-called shallowish connection charging). Similar to transmission use of system charges (TNUoS), distribution use of system charges (DNUoS) are charged to generators and loads: generators pay per MW connected and loads pay per MW and per MWh. Most of the revenue of a distribution company is collected from loads and this is disproportionately paid by households. As we noted earlier there are strong incentive payments for quality of service (e.g. reducing customer minutes lost) and these can substantially increase the rate of return on a distribution company’s assets. Over recent years, new generation is increasingly connecting to distribution grid (at 132/33 kV and below). Since 2011 around 13 GW of solar (all of it connected to the distribution grid) has been connected in Great Britain against peak demand of 54 GW. The charging methodology is substantially based on per MWh charges for smaller customers who are not half-hourly metered. There have been worries that this does not reflect the fixed costs of the network and hence over-incentivises

4  How Industrial Electricity Prices Are Determined… 

203

self-­generation (and storage).60 However this might be offset by a managed rise in electric vehicle charging within the distribution network that does not substantially add to the peak system requirements but makes more use of the existing distribution network.61 The future roles of transmission owners, the system operator and the distribution network companies are evolving as distributed generation (DG) increases and demand falls on the national transmission system. Who should have balancing responsibility and how should they fulfil it? There is a potential for distribution companies and third parties (such as customers and microgrids) to take more responsibility in system balancing and other traditional functions of the transmission network and its system operator. Balancing the system can be met by market-based solutions or via regulated assets (e.g. should a storage facility be a commercial or regulated asset?). The distribution system has traditionally been a passive network; however, the rise of DG means that it is becoming more active. This has led to reactive power (voltage) issues in parts of network, which could be procured locally or mitigated by the action of the distribution company. The benefits of any new arrangement need to be proven for customers and some ongoing innovation projects are trialling novel solutions to this.62 The lessons for China from the experience in Great Britain can be summed up as follows. Distribution pricing is an important component of overall electricity costs and incentive regulation can deliver impressive results. How overall charges are apportioned is very important and potentially highly distortionary in a more active network world. Therefore, there is a need to think carefully about how to deliver locational incentives. Recovering network fixed costs is a major issue for distribution networks, especially for a system where passive residential consumers are not shouldering a significant share of network costs at the moment. Technological developments will heighten tariff methodology issues everywhere, including in China.  See Pollitt (2018b) for a discussion.  See Küfeoğlu and Pollitt (2018) for some analysis of the impact of EVs on who pays residential distribution charges in GB. 62  See, for example, National Grid and UKPN’s Power Potential Project. The background issues to this project are discussed in Anaya and Pollitt (2018). 60 61

204 

M. G. Pollitt

4.10 Environmental Levies and Taxes There are a number of important taxes and levies which are part of the industrial price of electricity in GB.  These are a portion of the main renewables support schemes, namely, the Renewables Obligation (RO), the feed in tariffs (FITs) for smaller generators and contracts-for-­ differences (CFDs) (which will include nuclear power eventually). In addition, industrial customers contribute to the Hydro-Benefit Scheme, to support consumers in northern Scotland, and pay climate policyinspired energy efficiency charges including the climate change levy/carbon reduction commitment (CCL/CRC) and carbon pricing via the impact on generation prices of EU Emissions Trading Scheme (EU ETS) and carbon taxes in the form of the domestic carbon price support (CPS). We discuss each element in turn. The RO Scheme is a tradable green certificate scheme. Suppliers/retailers must present renewable obligation certificates (ROCs) for a percentage of sales. Renewable generators must be registered on the Renewables and CHP register at Ofgem to be awarded ROCs.63 For instance, in 2014–15, 71.3 million ROCs presented for 1 MWh each, which represented 99.1% of the total obligation on suppliers. The administratively set buyout price was £43.30 per ROC. The buyout price set the price for the ROCs presented. The buyout revenue is recycled to actual suppliers of ROCs, meaning that each ROC was worth £43.65 (recycle value was £0.35 plus the £43.30 buyout price) to a renewable generator.64 There is an 85% exemption from paying towards the ROC scheme for energy intensive users, but normally it is recovered by retailers per MWh on all loads.65 The RO scheme closed in 2017 to new generators but it is currently the most significant renewable support scheme financially. Small-scale FIT payments offer fixed prices per MWh to generators of different sizes (but usually less than 5 MW). The technologies effected  See https://www.renewablesandchp.ofgem.gov.uk/Public/ReportManager.aspx?ReportVisibility= 1&ReportCategory=0 64  See https://www.ofgem.gov.uk/environmental-programmes/renewables-obligation-ro 65  See Grubb and Drummond (2018). 63

4  How Industrial Electricity Prices Are Determined… 

205

include wind, solar, hydro and anaerobic digestion.66 These were initially very generous for solar, given the rapidly falling price of PV. CFDs are now the main way by which the UK government supports renewables (and new nuclear). It is set to become significant as projects financed by CFD contracts are completed. There have been auctions for CFD contracts in February 2015, August 2017 and August 2019, which have delivered significant quantities of lower price bids than the previous—interim—administrative CFD prices. In the first auction, winning onshore wind bids were 17% lower than the administrative CFD price and winning offshore wind bids were 18% lower than the previous administrative price.67 In the second auction winning bids fell again, with offshore wind projects winning at a price of £57.50/MWh for delivery in 2022/23.68 This was against an administrative price of £140/MWh prior to the first auction, only two and half years earlier. The third auction saw offshore wind prices fall again to £41.61/MWh for delivery in 2024/25. The Hydro-Benefit Scheme is an interesting cross-subsidy paid by all customers in GB to reduce the high costs of electricity distribution in the region with the lowest population density. Following liberalisation in 1990, the introduction of the wholesale power market threatened to unwind the internal cross-subsidy (within an integrated utility) between the low cost of generation and high cost of distribution in the Scottish Hydro area, north of Scotland. Initially the Hydro-Benefit Scheme taxed the hydrogeneration and subsidised the distribution charges in the Scottish Hydro area. Later the high distribution cost was covered by levy on all consumption across GB—via the Hydro-Benefit Replacement Scheme.69 Industrial customers can be subject to two energy efficiency taxes, which although they are nominally related to climate policy do not tax carbon directly but energy use. The climate change levy (CCL) was set at  See Helm (2017, p. 101).  See https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_ data/file/407465/Breakdown_information_on_CFD_auctions.pdf 68  See https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_ data/file/643560/CFD_allocation_round_2_outcome_FINAL.pdf 69  See DECC (2015), https://assets.publishing.service.gov.uk/government/uploads/system/uploads/ attachment_data/file/488271/decc_consultation_hydro_benefit_review_22_dec_15__2_.pdf 66 67

206 

M. G. Pollitt

£5.83/MWh in April 2018. It is charged to large energy-intensive users and is subject to a 90% rebate if a voluntary climate change agreement is in place. The carbon reduction commitment (CRC) was set at £17.20–18.30/tonne CO2 in 2018/19 on larger commercial users of electricity to encourage investment in energy management. It is calculated on the deemed carbon content of grid supplied electricity. It has been abolished from 2019. Carbon pricing has a significant impact on the industrial electricity price. This happens in two ways: via the participation of the UK electricity sector in the EU ETS and the additional imposition of a carbon tax on fossil fuels used in electricity generation in the UK, via the carbon price support (CPS). The CPS effectively increases the price of carbon emissions from the electricity sector in the UK above that in other member countries of the EU ETS. The CPS is part of the carbon price floor (CPF) which sets a target price for the combined EU ETS and CPS price in the UK. It began in April 2013 with a target CPF CO2 price of £30/tonne (in 2009 terms)—forward EUA price + CPS—by 2020 (possibly £70/tonne by 2030). However, the CPS is now capped at £18/tonne CO2 (now binding). The CPS directly impacts the wholesale price via raising the price of marginal fossil generation. EUA price currently is £14.08/tonne CO2 (13/07/18). By 2017, the impact of £18/tonne CPS was enough to push much of the remaining coal-fired generation off the system.70 The combined impact of these levies and taxes on the industrial price of electricity is substantial in Great Britain. A key question for China is the extent to which industrial electricity customers can and should be subject to payments for low carbon generation, energy efficiency and carbon pricing. Some other countries, such as Germany, have exempted much of their industry from bearing the costs of government policy in the electricity sector. This is only possible in systems where industry is a relatively small share of total electricity demand. This is not the case in China. It is right that all electricity consumers pay the true cost of electricity and this includes charges which reflect the externality cost of carbon emissions from power plants or the local environmental benefit of cleaner technologies. However, it remains an open question as to whether  See Wilson and Staffell (2018).

70

4  How Industrial Electricity Prices Are Determined… 

207

some of the cost of the energy transition should be shifted from electricity consumers onto general taxation because the current relatively high cost of renewables is a function of its technological immaturity and hence there is a wider public benefit from the learning by doing effect of subsidy.71 Energy efficiency policies are wider than just electricity use and hence it is worth thinking about whether payment for these policies are fairly targeted on electricity users, especially when it may be poorer electricity customers that end up paying disproportionately for them.

4.11 O  verall Lessons on Price Determination for China from Great Britain Following the theory of the spot pricing of electricity due to Schweppe and colleagues,72 the price for every industrial customer should vary by location, time, quantity and willingness to accept interruption. However, in the real world of liberalised markets there is much less bill variation than the underlying price components would suggest as final customers value certainty in pricing. In general, the focus in a liberalised market is on what the customer is getting for their money and away from the producer, except in the sense that producer needs a fair return on capital. In an initially profitable system—such as in China—reform should be about rebalancing the electricity system away from producer to consumer interests, that is, from inefficient costs and high profits towards cheaper, cleaner and more reliable electricity supply. There is a key role for the profit motive in a liberalised market as a guide to decision making. Transparency on price components is important for promoting better regulation and more competition. Wholesale power and ancillary services costs are reduced over the longer run by the use of wholesale spot markets to guide both short-term dispatch and long-term investment in fossil fuel power plants. Transmission and distribution charges are an important component of costs; even in the UK these are 20% of the industrial price (where generation costs are 33%) 71 72

 See Newbery (2017).  See Bohn et al. (1984).

208 

M. G. Pollitt

and incentive regulation of network charges can bring large improvements in both cost efficiency and network quality. In China, there is a need to understand and expand the role of retailers by separating them fully from distribution. State Grid Company of China and China Southern Grid (SGCC and CSG) provincial retail should be fully legally unbundled from distribution and allowed to compete nationally for retail customers. Competition in generation and retail needs to be effectively overseen and regulated by both the regulator (the NEA) and the Anti-­ Monopoly authorities as pressure to consolidate the sector and undesirable price discrimination is likely. In China, there is a need for a focus on the big picture (e.g. how much have aggregate prices/efficiency/profits changed) rather than just the detail (e.g. zonal vs. nodal pricing, central vs. self-dispatch). The aim should be to stop the power sector being subject to purchasing requirements for domestic technology and domestic natural resources (in GB the electricity industry eventually escaped from its historic commitment to buying expensive British coal). In China, local taxation and non-­ externality related charges can distort production choices and impose unnecessary industrial policy costs on other industrial electricity customers. Instead the power sector’s key role should be understood to be in promoting development in the wider economy by efficient (and fully cost reflective) pricing. It is important to produce electricity efficiently and use taxation to drive up the price to promote energy efficiency and decarbonisation, rather than letting incumbents justify high prices on grounds of energy efficiency. Challenges remain for all countries, including China, in the future development of the power sector, with the rise of new distributed energy technologies. The current electricity system is characterised by high fixed costs which should be recovered. It is difficult to prevent behind-the-­meter investments to avoid paying towards these fixed costs. This suggests there may be a need to lift some electricity system costs to general taxation (e.g. energy R+D, energy efficiency measures). In China, as the growth in the number of kWhs distributed slows attention to fixed costs will increase. More competition and better network regulation will lower profit margins at home, to the benefit of consumers and the discouragement of wasteful investment abroad, and it will reduce concerns about private/foreign ownership in the electricity sector, as has happened in the UK.

4  How Industrial Electricity Prices Are Determined… 

209

References Anaya, K., & Pollitt, M. (2017). Regulating the electricity system operator: Lessons for Great Britain from around the world. EPRG working paper, No. 1718. Cambridge: University of Cambridge. Anaya, K., & Pollitt, M. (2018). Reactive power procurement: Lessons from three leading countries. EPRG working paper, No.1829. Cambridge: University of Cambridge. BEIS. (2018). Digest of UK energy statistics 2018. London: Department of Business, Energy and Industrial Strategy. Bohn, R. E., Caramanis, M. C., & Schweppe, F. C. (1984). Optimal pricing in electrical networks over space and time. RAND Journal of Economics, 15(3), 360–376. CMA. (2016a). Energy market investigation. Appendix 7.1: Liquidity. London: Competition and Markets Authority. CMA. (2016b). Energy market investigation. Appendix 5.2: Locational pricing in the electricity market in Great Britain. London: Competition and Markets Authority. CMA. (2017). Market studies and market investigations: Supplemental guidance on the CMA’s approach. London: Competition and Markets Authority. DECC. (2009). Digest of United Kingdom energy statistics: 60th anniversary. London: Department of Energy and Climate Change. DECC. (2011). Digest of United Kingdom energy statistics 2011. London: TSO. DECC. (2015). Hydro-benefit replacement scheme and common tariff obligation, three year review of statutory schemes: Consultation. London: Department of Energy and Climate Change. Domah, P. D., & Pollitt, M. G. (2001). The restructuring and privatisation of the regional electricity companies in England and Wales: A social cost benefit analysis. Fiscal Studies, 22(1), 107–146. Evans, J., & Green, R. (2003). Why did electricity prices fall after 1998? Cambridge Working Papers in Economics No.0326. Cambridge: University of Cambridge. Greve, T., Teng, F., Pollitt, M. G., & Strbac, G. (2018). A system operator’s utility function for the frequency response market. Applied Energy, 231, 562–569. Grubb, M., & Drummond, P. (2018, February). Industrial electricity prices: Competitiveness in a low carbon world. Report Commissioned by the Aldersgate Group. London: UCL Energy Institute. Helm, D. (2017). Cost of energy review. https://assets.publishing.service.gov.uk/ government/uploads/system/uploads/attachment_data/file/654902/Cost_ of_Energy_Review.pdf

210 

M. G. Pollitt

Henney, A. (1994). A study of the privatisation of the electricity supply industry in England and Wales. London: EEE Ltd. Henney, A. (2011). The British electric industry 1990–2010: The rise and demise of competition. http://alexhenney.com/order_the_book.htm Hirst, D. (2018, January 8). The carbon price floor and the price support mechanism. Briefing paper No. 05927. London: House of Commons Library. Hogan, W.  W. (1992). Contract networks for electric power transmission. Journal of Regulatory Economics, 4(3), 211–242. Jamasb, T., & Pollitt, M. (2005). Electricity market reform in the European Union: Review of progress toward liberalization and integration. The Energy Journal, Special Issue on European Electricity Liberalisation, 11–41. Jamasb, T., & Pollitt, M. (2007). Incentive regulation of electricity distribution networks: Lessons of experience from Britain. Energy Policy, 35(12), 6163–6187. Küfeoğlu, S., Pollitt, M. (2018). The impact of PVs and EVs on domestic electricity network charges: A case study from Great Britain, EPRG working paper, No. 1814. Cambridge: University of Cambridge. Mansur, E., & White, M. (2012). Market organization and efficiency in electricity markets. mimeo. https://www.dartmouth.edu/~mansur/papers/mansur_ white_pjmaep.pdf MMC. (1996a). National Power plc and Southern Electric plc: A report on the proposed merger, Monopolies and Mergers Commission, Cm 3230. London: HMSO. MMC. (1996b). PowerGen plc for Midland Electricity plc, A Report on the proposed merger. Monopolies and Mergers Commission, Cm 3231. London: HMSO. National Grid. (2011, September 20). Managing Intermittent and Inflexible generation in the Balancing Mechanism. National Grid. National Grid. (2018, January). Final TNUoS Tariffs 2018/19. National Grid. National Grid ESO. (2019). Power responsive: Demand side flexibility annual report 2018. National Grid ESO. Newbery, D. M. G. (2000). Privatization, restructuring, and regulation of network utilities: The Walras-Pareto lectures. Cambridge, MA: MIT Press. Newbery, D. M. G. (2005). Electricity liberalization in Britain: The quest for a satisfactory wholesale market design. The Energy Journal, 26.((Special Issue: European Electricity Liberalisation), 43–70. Newbery, D. (2012). Contracting for wind generation. Economics of Energy and Environmental Policy, 1(2), 19–36.

4  How Industrial Electricity Prices Are Determined… 

211

Newbery, D. (2017). How to judge whether supporting solar PV is justified. EPRG working paper, No. 1706. Cambridge: University of Cambridge. Newbery, D. M. G., & Pollitt, M. G. (1997). Restructuring and privatisation of the CEGB – Was it worth it? Journal of Industrial Economics, 45(3), 269–304. Nillesen, P. H. L., & Pollitt, M. G. (2011). Ownership unbundling in electricity distribution: Empirical evidence from New Zealand. Review of Industrial Organization, 38(1), 61–93. Offer. (1991, December). Pool price enquiry. Birmingham: Office of Electricity Regulation. Offer. (1992). Report on constrained-on plant. Birmingham: Office of Electricity Regulation. Offer. (1993, July). Pool price statement. Birmingham: Office of Electricity Regulation. Offer. (1998). Review of electricity trading arrangements: Proposals. Birmingham: Office of Electricity Regulation. Ofgem. (2009a). Electricity distribution price control initial proposals, incentives and obligations. Ref. 93/09. London: Ofgem. Ofgem. (2009b). Electricity distribution price control final proposals. Ref. 144/09. London: Ofgem. Ofgem. (2009c). Regulating energy networks for the future: RPI-X@20 principles, process and issues. London: Ofgem. Ofgem. (2014a). RIIO-ED1: Final determinations for the slow track electricity distribution companies, detailed figures by company. London: Ofgem. Ofgem. (2014b). RIIO-ED1 draft determinations for slow track electricity distribution companies. London: Ofgem. Ofgem. (2016). Wholesale power market liquidity: Annual report 2016. London: Ofgem. Ofgem. (2018a). State of the energy market report 2018. London: Ofgem. Ofgem. (2018b). Ofgem annual report and accounts 2017–18. London: Ofgem. Ofgem. (2018c). Consumer impact report financial year 2017–18. London: Ofgem. Ofgem. (2018d). The electricity system operator regulatory and incentives framework from April 2018. London: Ofgem. Onaiwu, E. (2009). How does bilateral trading differ from electricity pool trading?. University of Dundee. https://archive.uea.ac.uk/~e680/energy/energy_links/ electricity/…. 27 Sept 2012. PA Consulting Group. (2016, May). OFGEM – Aggregators barriers and external impacts. London: PA Consulting Group.

212 

M. G. Pollitt

Pollitt, M. (1999). The survey of the liberalization of public enterprises in the UK since 1979. In M. Kagami & M. Tsuji (Eds.), Privatization, deregulation and institutional framework (pp.  120–169). Institute of Developing Economies: Tokyo. Pollitt, M. G. (2008). The future of electricity (and gas) regulation in a low-­ carbon policy world. The Energy Journal, 29(S2), 63–94. Pollitt, M. (2012a). The role of policy in energy transitions: Lessons from the energy liberalisation era. Energy Policy, 50(November), 128–137. Pollitt, M. G. (2012b). Lessons from the history of independent system operators in the energy sector. Energy Policy, 47(August), 32–48. https://doi. org/10.1016/j.enpol.2012.04.007. Pollitt, M. (2018a), The European single market in electricity: An economic assessment. Energy Policy Research Group working papers No. 1815. Cambridge: University of Cambridge. Pollitt, M. G. (2018b). Electricity network charging in the presence of distributed energy resources: Principles, problems and solutions. Economics of Energy and Environmental Policy, 7(1), 89–103. Pollitt, M., & Haney, A. B. (2014). Dismantling a competitive retail electricity market: Residential market reforms in Great Britain. The Electricity Journal, 27(1), 66–73. Pollitt, M., & Stern, J. (2011). Human resource constraints for electricity regulation in developing countries: Developments since 2001. Utilities Policy, 19(2), 53–60. Sioshansi, R., Oren, S., & O’Neill, R. (2008). The cost of anarchy in self-­ commitment-­based electricity markets. In F. P. Sioshansi (Ed.), Competitive electricity markets: Design, implementation and performance (pp.  245–266). Amsterdam: Elsevier. Stoft, S. (2002). Power system economics: Designing markets for electricity. Piscataway: Wiley-IEEE Press. Sweeney, J. L. (2002). The California electricity crisis. Hoover Institution Press. Taylor, S. (2007). Privatization and financial collapse in the nuclear industry – The origins and causes of the British Energy crisis of 2002. London: Routledge. Taylor, S. (2016). The fall and rise of nuclear power in Britain: A history. Cambridge: UIT Cambridge. Vona, F., & Nicolli, F. (2014). Energy market liberalization and renewable energy policies in OECD countries. IEB working paper. Wilson, I. A. G., & Staffell, I. (2018). Rapid fuel switching from coal to natural gas through effective carbon pricing. Nature Energy, 3, 365–372.

4  How Industrial Electricity Prices Are Determined… 

213

Additional Useful Resources on UK Electricity Reform Oral History of Electricity Privatisation: British Library archive of key players: This online archive is extensive with many recordings from different players; some are shown below:   •  Cecil Parkinson – Secretary of State for Energy   – http://sounds.bl.uk/Oral-history/Industry-water-steel-and-energy/021MC1495X0021XX-0001V0   •   John Wakeham – Secretary of State of Energy   – http://sounds.bl.uk/Oral-history/Industry-water-steel-and-energy/021MC1495X0048XX-0001V0   •  William Rickett – Civil Servant involved with privatisation   – http://sounds.bl.uk/Oral-history/Industry-water-steel-and-energy/021MC1495X0033XX-0004V0   •  Brian Pomeroy – Advisor on Electricity Privatisation   – http://sounds.bl.uk/Oral-history/Industry-water-steel-and-energy/021MC1495X0048XX-0001V0   •  Fiona Woolf – Advisor on Electricity Privatisation   – http://sounds.bl.uk/Oral-history/Industry-water-steel-and-energy/021MC1495X0047XX-0001V0

5 Prospects for Reform of China’s Electric Power Sector

This final chapter seeks to summarise some of the key issues emerging from the previous chapters on the prospects for the reform of the Chinese electric power sector. In Sect. 5.1 we begin with some of the high-level messages from the previous four chapters. In Sect. 5.2 we comment on recent developments with the provincial power reform pilots across China. We then move on and make some suggestions for immediate next steps in Sect. 5.3. In Sect. 5.4 we discuss the fundamental issues that are raised in China by power sector reform. In Sect. 5.5 we end with some key messages for different stakeholder groups within the power sector, namely, policymakers, regulators, generators, retailers and the grid companies (specifically, State Grid Company of China [SGCC] and China Southern Grid [CSG]).

The author would like to thank Hao Chen for his help with the data in this chapter.

© The Author(s) 2020 M. G. Pollitt, Reforming the Chinese Electricity Supply Sector, https://doi.org/10.1007/978-3-030-39462-2_5

215

216 

M. G. Pollitt

5.1 High-Level Messages from Previous Chapters Chapter 1 highlighted the sheer scale and scope of the Chinese power sector. This is the world’s largest power sector on every dimension. It is getting bigger with the Chinese economy and the likelihood must surely be that Chinese electricity companies (and associated firms) will become more significant in the global power sector, at least to the extent that they can compete in international markets. A key motivation for reforming the domestic power sector is that without a competitive internal market it is difficult to see how Chinese electricity firms can compete in sophisticated power markets in the US, Europe and Australasia in electricity consumer-oriented technologies. International interest in the Chinese power market is motivated by its impact on the global environment, its key role in supplying global power equipment and its impact on the future development of the technology of the power sector. Chapter 2 made a number of suggestions for the improvement in the efficient operation of the Chinese power sector based on international experience of power market reform. Prior to the start of the 2015 reform Chinese industrial power prices were relatively high, relative to underlying fuel costs. There was scope for significant reductions in the price of industrial power prices if the sector was incentivised to deliver efficiency improvements (and reductions in margins) to consumers. These price reductions could be delivered at the same time as reducing emissions from the Chinese power sector, creating a further social benefit. We discussed how price reductions might be delivered by a combination of introducing merit order dispatch of power plants, incentive regulation of network charges, reducing the incentives to overinvest in the power system and rebalancing power prices towards domestic customers, who are currently underpaying for power. Chapter 3 focused on what was actually happening within one leading province (Guangdong) with power market reform. We discussed the extent to which the introduction of annual, monthly and day-ahead power markets was proceeding in line with international experience. We noted the massive increase in the number of new retailers in the

5  Prospects for Reform of China’s Electric Power Sector 

217

Guangdong market and the impact that power market reform was having on industrial prices, for customers inside and outside the power market. We observed that there has been a large price drop in the wholesale power market and that regulated network charges have come down. This suggests a positive experience so far from introducing wholesale market competition and incentive regulation of network charges. However significant issues remain and progress towards the full implementation of a power market which functions in line with successful experience of reformed power markets globally is some way off. Chapter 4 highlighted how industrial power prices are determined within a typical European power market, namely the UK.  We discussed each of the elements of the industrial price. We showed that for the UK in 2016 only around 40% of the final price that industrial customers actually pay is determined by the ‘market’ as ‘pure’ wholesale generation plus retail margin costs. Twenty per cent was network charges and system operator costs, which were mostly the result of the operation of incentive regulation of natural monopolies. The remaining 40% was government levies and taxes, largely relating to carbon pricing (arising from both the European Emissions Trading Scheme [EU ETS] and the UK’s own carbon tax on fossil fuels for power generation) and renewable support schemes. Thus the influence of the power market, even on the industrial price, is limited. This emphasises the key role for the regulator in overseeing incentive regulation of network charges and for government (which, in China, might be at the central, provincial or municipal level) in determining industrial and tax policy towards the power sector.

5.2 R  ecent Developments on Power Sector Reform in China Power sector reform in China is a massive undertaking. The power sector employs four million people. The scope for efficiency improvement is high with China’s electricity sector employing significantly more

218 

M. G. Pollitt

employees per TWh distributed than in the US and having a proportion of managers three times higher than in the US.1 While slightly lower average annual GDP growth (below 7% p.a.) may be ‘the new normal’, power sector demand will likely continue to grow strongly. Reforming the power sector needs to be done while still satisfying the growing demand for power. This is in sharp contrast to the US, Europe and Australasia where power demand peaked more than a decade ago and is currently not showing trend growth (see Sioshansi 2016). There has been a significant amount of activity around power sector reform in China. There has been significant institutional development. By the end of 2017 there were around 7000 retailers across China registered to participate in provincial electricity markets.2 These were all start-­ ups, even though some of them are backed by well-capitalised existing firms. Thirty-five power exchanges (33 at provincial level and 2 at regional level in Beijing and Guangzhou) have been created (by January 2018) as subsidiary companies of the grid companies.3 Grid company dispatch centres have had to adjust to the need to pay more attention to market outcomes in the annual and monthly markets (and soon to day-ahead markets). The National Energy Administration (NEA) and National Development and Reform Commission (NDRC)—the branches of government overseeing power market reform at both the national and provincial level—have had to invest significantly in learning about power markets and facilitating their creation. The NEA and NDRC have continued to actively promote reform most recently (in July 2019) with the document ‘Opinions on Deepening the Electricity Spot Market Construction in Pilot Areas’.4 Where monthly and annual power markets have been implemented there would appear to have been significant price reductions relative to the regulated generation price for participating customers, as we saw in the case of Guangdong in Chap. 3. It would also appear that regulated industrial prices have also fallen due to a combination of the  See Rawski (2019, p. 350).  See Alva and Li (2018, p. 21). 3  See Alva and Li (2018, p. 38). 4  NDRC (2019), Opinions on Deepening the Electricity Spot Market Construction in Pilot Areas. Available at: http://www.ndrc.gov.cn/gzdt/201908/t20190807_943964.html 1 2

5  Prospects for Reform of China’s Electric Power Sector 

219

implementation of the power market, the regulation of network charges and a reduction of government taxes on power. The US industrial electricity price was 2.4% lower in 2018 than it was in 2014 in nominal USD.5 By contrast the regulated price of industrial electricity in Guangzhou was reportedly 8% lower in RMB in 2018 than in 2014 (and 14% lower in USD).6 For industrial customers in the power market (before any payment of retailer margin) the nominal reductions in 2018 increased to 17% in RMB (and 23% in USD),7 though market discounts were lower in 2019. This indicates that the reform is delivering on the high-level objective we outlined in Chap. 1. Since 2017 there have been eight designated power market pilot areas: Inner Mongolia, Fujian, Guangdong, Shandong, Zhejiang, Sichuan, Gansu and Shanxi. Each of these provinces exhibits different economic circumstances as shown in Table 5.1, where the provinces vary significantly in average income and in unit electricity price, with the richer provinces exhibiting significantly higher electricity prices (up to 40% higher). The provinces are also of varying electrical sizes and with different mixes of imports and exports, as shown in Table 5.2. It is interesting that provinces with relatively low prices in Table 5.1 are also significant net exporters (Inter Mongolia, Sichuan and Shanxi), while provinces with significant net imports are those with higher prices (Guangdong, Shandong and Zhejiang). The pilot areas include two provinces with much more limited trading (Fujian and Gansu). The provinces have very different sources of local generation (as shown in Table 5.3), with Sichuan and Gansu with much lower shares of thermal (mostly coal) generation than the other provinces. All of the pilot provinces have commenced with an annual contract power market. However the pilot provinces are moving at different speeds in terms of their further progress with reform. This can be seen in terms of the date at which they have implemented the monthly bidding market (which we discussed for Guangdong in Chap. 3). Indeed, as of October  Source: EIA (2019).  35 kV+ customer in Guangzhou. Source: CEIC. US $1 = 6.1428 CNY in 2014 and US $1 = 6.62 CNY in 2018. Note: the figures are difficult to get on a consistent basis. 7  Assuming the 2018 annual bilateral contract discount of 0.0782 RMB/kWh. 5 6

68,302 91,197 86,412 76,267 98,643 48,883 31,336

25.34 39.41 113.46 100.47 57.37 83.41 26.37

Population (million)

0.4489 0.5802 0.6084 0.6206 0.6644 0.5774 0.4632

Regulated electricity price RMB/ kWh 10 kV 0.4379 0.5602 0.5834 0.6056 0.6344 0.5574 0.4532

Regulated electricity price RMB/kWh 35 kV 0.4279 0.5402 0.5834 0.5906 0.6124 0.5374 0.4432

Regulated electricity price RMB/kWh 110 kV

0.4234 0.5202 0.5584 0.5756 0.6074 0.5174 0.4342

Regulated electricity price RMB/kWh 220 kV

Source and notes: NBS (2018) for GDP per capita and population. Price data are from China electric power knowledge database as of April 1, 2018. Average price of west and east Inner Mongolia

Inner Mongolia Fujian Guangdong Shandong Zhejiang Sichuan Gansu

GDP per capita RMB

Table 5.1  Characteristics of the pilot provinces in 2017

220  M. G. Pollitt

221

5  Prospects for Reform of China’s Electric Power Sector  Table 5.2  Provincial electricity balances in pilot areas in 2017

Inner Mongolia Fujian Guangdong Shandong Zhejiang Sichuan Gansu Shanxi

Consumption TWh

Production TWh

289.2 211.3 595.9 543.0 419.3 220.5 116.4 199.1

442.4 218.6 434.8 486.0 334.8 356.9 134.2 276.6

Net imports TWh

Net exports TWh 153.2 7.3

161.1 57.0 84.5 136.4 17.8 77.5

Source: NBS (2018)

Table 5.3  Electricity production mix in pilot provinces in 2017 Hydro (%) Thermal (%) Nuclear (%) Wind (%) Solar (%) Inner Mongolia Fujian Guangdong Shandong Zhejiang Sichuan Gansu Shanxi

2.05 23.35 13.55 0.86 13.04 79.35 17.38 3.02

69.09 54.92 70.82 82.31 68.93 17.10 41.22 78.86

0.00 15.56 9.54 0.00 7.38 0.00 0.00 0.00

22.58 4.50 3.05 8.45 1.49 2.16 25.67 10.80

6.28 1.64 3.03 8.38 9.15 1.39 15.74 7.31

Source: China Electric Power Yearbook 2018

2019, Zhejiang had only recently started trialling a monthly bidding market which would allow its registered retailers to start trading. Since 2014 it had allowed direct trading between generators and customers of annual contracts. In 2017 there were also very different degrees of progress in terms of the extent of demand covered by the power market (both annual and monthly). This ranged from 20% in Fujian to 50% in Inner Mongolia (Table 5.4). As noted above, power markets have generally seen significant reductions in reported prices, relative to the regulated power price within the province and seen significant increases in the number of retailers. The reported profitability of generators, increasingly exposed to the wholesale

222 

M. G. Pollitt

Table 5.4  Progress with power market implementation

Inner Mongolia Fujian Guangdong Shandong Zhejiang Sichuan Gansu Shanxi

Start of monthly bidding market

Share of demand in power market 2017 (%)

Apr-17 May-18 Feb-17 Jul-17 Has not started Jun-17 Dec-17 Apr-17

50.90 20.30 29.10 37.30 36.10 30.70 33.60 34.80

Source: News websites and www.cec.org.cn for share of demand in power market. Inner Mongolia share of demand in power market is a simple average of east and west

power market, appears to have declined significantly,8 since the start of power market reform. The major network companies, SGCC and CSG, who have been subjected to significant reductions in their charges, have also seen their profitability decline.9 Power market reform is therefore having the effects that we have seen in many countries across the world, reducing prices to electricity customers while putting pressure on incumbents to reduce their own costs or face reductions in their profit margins. It is important to note that power market reform is proceeding in other provinces as well. Jiangsu (China’s second richest province, only narrowly behind Guangdong in terms of contribution to GDP in 2017) had 212 registered retailers at the end of 2018 and monthly and annual contract price reductions of 0.02 RMB per kWh relative to the regulated price. This compares with Zhejiang where there are 300 registered retailers (none of which are trading in the market) and annual contract market price reductions of around 0.03 RMB per kWh relative to the regulated price in 2018. As of mid-2019, Jiangsu had plans to implement a  Profits at four of the Big five retailers—Huaneng, Huadian, State Power Investments and Datang—fell from $2461 m in 2016 to $967 m in 2019 (Source: Fortune Global 500). Guodian had merged with another major energy company (Shenhua) in the interim, making it difficult to compare its profits. 9  Profits at SGCC and CSG were $8174 m and $1782 m in 2019 against $10,201 m and $2223 m in 2016 (Source: Fortune Global 500). 8

5  Prospects for Reform of China’s Electric Power Sector 

223

competitive spot market, similar to that being established in Guangdong, in 2020. Thus it is quite possible that some non-pilot markets will overtake the pilot market provinces in terms of their progress with implementing power market reform. While there has been significant progress with annual and monthly contracting, the move from that to a fully integrated day-ahead and real-­ time market has been more difficult (as we saw in Chap. 3 for Guangdong). This has been because it has taken some time to implement the required software and integrate this with dispatch decisions. All of the pilot provinces, and those other provinces with operational contract markets, have organised their annual and monthly contract markets through power exchanges. However the move to a day-ahead market integrated with dispatch decisions requires the leadership of the dispatch centres of the power grid companies—these are separate organisations (even though both entities are currently wholly owned subsidiaries of the provincial grid companies in their relevant areas). Trials of the day-ahead (spot) markets are at advanced stages in Guangdong and Zhejiang. Zhejiang ran an international tender in 2017 for the software to implement its spot market. This tender was won by PJM (the largest independent system operation in the US) and the software has now been written to run its spot market and simulations have been done using the software. Guangdong has also implemented the software needed for a PJM-type nodal pricing regime, using Chinese software developers. It has trialled its spot market with real money on a number of trial trading days in 2019. However the decision to run the spot market continuously has not (as of October 2019) been taken in either province. This is partly because the implications of full merit order dispatch are significant. Power prices on the day will be subject to potentially significant intra-day and intra-provincial price variation. This will impact on plant running decisions. It will also highlight the impact of grid decisions on the nodal (and provincial) prices. This is not just about generator and customer actions. The grid company makes market sensitive actions, for instance when it takes lines out of service. Procedures for notification of the market need to be in place and non-discriminatory and the grid

224 

M. G. Pollitt

company needs to minimise the market cost of its actions, minimising loss of capacity and thinking more carefully about the sequencing of its maintenance and upgrade activities. Full nodal spot pricing also allows the exploitation of short-run and local market power. Individual generator units can now exploit their monopoly positions behind constraints and their availability at peak times. This tests the procedures for monitoring the exercise of market power and for distinguishing legitimate reasons for high price bids or non-availability from the illegitimate exploitation of market power. Thus full operation of the day-ahead market, with its implications for dispatch, grid actions and regulatory oversight, exposes the market to even sharper stress tests. Not surprisingly, there has been much less progress with ancillary services markets, with the most promising area of this being some trading of coal generator transmission rights in order to facilitate the utilisation of renewable generation.10 Guangdong has however trialled a day-ahead spot market for frequency response. At the moment much of the discussion of power market reform in China has focussed on the technical details of implementation, that is how to put in place a full set of electricity markets and in particular how to work towards a spot market. However once these technical details have been worked out and the markets are fully implemented there will be a shift of emphasis towards how these new arrangements are actually delivering for customers. Customers in China’s power markets have so far been offered lower prices via their annual and monthly contracts, but the question for them is what additional benefit they will see from spot markets.

5.3 Suggestions for Next Steps We make a number of constructive suggestions for next steps based on international experience of reform.

 See Alva and Li (2018, pp. 43–44) who discuss the North-east ancillary services market covering Liaoning, Jilin and Heilongjiang provinces. 10

5  Prospects for Reform of China’s Electric Power Sector 

225

5.3.1 Improving Regulatory Capacity Electricity markets are sophisticated market arrangements with ample opportunity for anti-competitive gaming. This has been amply demonstrated by the experience of many markets across the world. What is needed is proper monitoring of behaviour, rapid intervention and strong competition policy to deal with abuses of the market. A competitive market will also result in some firms, both retailers and potentially generators going out of business. Thus it also requires good procedures to handle financial failure on the part of market participants. Electricity markets require sophisticated regulation and competition policy. China needs to put in place effective institutions of regulation and competition policy in order for power market reform not to be derailed by market power abuses or indeed the—largely positive—consequences of intense market competition. The transition to a fully functioning electricity market requires the transition to a regulatory system capable of matching the market with its own sophistication and ability to respond to emerging information on the nature of the market.

5.3.2 Improving Regulatory Reporting A key suggestion would be to improve regulatory reporting on market performance. There are several examples of good regular reporting on the state of the electricity market in the US, UK and in Australia. The Federal Energy Regulatory Commission (FERC), which oversees energy markets across the US, produces an annual State of the Markets Report and individual markets within the US produce their own detailed reports. One such example is the two-volume State of the Market Report for PJM (which stretches to more than 750 pages). In the UK, Ofgem produces an annual State of the Energy Market report for the Great Britain market, while the Australian Energy Regulator has a similar report for Australia. It is worth drawing attention to some of the interesting things that these reports highlight. FERC (2019) includes, inter alia, power price movements across the US, capacity market prices and sources of additions of new capacity. PJM publishes reviews (see Monitoring Analytics

226 

M. G. Pollitt

2019a, b) of the state of competition within each submarket (e.g. energy, capacity, scheduling reserve, synchronised reserve, frequency regulation). For each submarket the annual review assesses the market structure in the aggregate market, the local nodal markets, participant behaviour and market performance offering an evaluation of whether it was competitive, not competitive or partially competitive. The review offers an overall conclusion on whether the market design was effective, mixed or flawed. The PJM report also includes price data for the last ten years on the different market prices. The GB report includes analysis of generator profitability, trends in tariff offerings, customer surveys on the degree of engagement with the market, the evolution of average prices and various price components (Ofgem 2018). The Australian report monitors new investment and capacity withdrawals, distribution company revenues, measures of competition in generation (HHIs) and the pivotality of certain generators (AER 2018).

5.3.3 Promoting Learning from the Pilot Markets There is much scope for China to learn from the experiences of the different pilot markets (and those other provinces such as Jiangsu that are actively pursuing reform). This is best promoted by encouraging the publication of as much data as possible about the progress with reform. Market data should be freely available on public webpages for analysis and not confined to market participants. There should be encouragement to publish information on the emerging lessons from the individual pilots to promote learning across the provinces. One of the lessons from jurisdictions that have successfully promoted reform is that the promotion of transparency is a good way to ensure that problems with market design and market power are addressed more quickly. It is the role of the NEA to monitor power market reform in the different provinces. It should actively promote positive lessons from provinces that have gained a good understanding of how to implement power markets and also suggest remedies as problems have emerged in some provinces. For instance, Guangdong seems to have made good progress

5  Prospects for Reform of China’s Electric Power Sector 

227

with actually implementing monthly and spot markets, whereas Zhejiang has struggled to do so. Highlighting the contrasting experience of these two well-resourced provinces with similar starting points and intentions would seem to be important. FERC in the US played a role in promoting the learning from successful ISO markets in the US to other organised markets in the US. Similarly, the European Union built its target model for electricity market design within the EU, which emerged in 2004, on the lessons from Northern European states that successfully implemented power markets in the 1990s.11

5.3.4 P  utting All Generation and Demand in the Wholesale Market There is a need to complete the wholesale energy market in the sense that all generation and all demand needs to be in the market. One of the observations from all organised markets in the US, Europe, Australasia and South America is that all demand, including demand from price-­ regulated domestic customers, is in the wholesale market. This means that all generation and all demand is subject to competition and deepens the market. Thus incumbent retailers still purchase power on behalf of their regulated customers through the market. In China this is a large share of the market (almost 70% across China, and more in some provinces) and this is currently completely outside the trails for the wholesale power market. How to bring this into the market is a challenge as it would involve getting SGCC and CSG to act as retail participants in the existing annual and monthly contract markets. This would significantly increase competition in these markets, but immediately raise the question of the need to separate the retail divisions of SGCC and CSG from their network divisions, as has happened across other reformed power markets globally. This could be done by initially pursuing legal separation of networks from retailing, with a view to ownership unbundling at some point in the future in order to create more retail competition.

11

 See Pollitt (2019) on the development of the EU single electricity market.

228 

M. G. Pollitt

5.3.5 C  onsider Whole or Part Privatisation of One Large Generator The Chinese electricity sector is still dominated by large state- or provincially owned electricity companies. Generation in particular is largely government owned. Given the centrality of the profit motive in delivering rational responses to market prices, some consideration ought to be given to the whole or part privatisation of at least one of the big five generating companies. This would be in line with global experience that it is profitoriented generation companies that pursue the benefits of competition and have the incentive to respond fully to market price signals. It is difficult to see how a sector dominated by state-owned companies can exploit the opportunities provided by market prices. The privatisation of one generator would be a prudent, but significant, first step to demonstrating the possibilities of competition. Indeed China could privatise the smallest of the big five generators (by capacity) with a view to shaking up the behaviour of the others. The privatisation would not need to be 100%, simply enough to give effective control to private investors and properly align management incentives.

5.3.6 T  he Creation of Genuine Interprovincial Market Should Be Done in Stages The priority for power sector reform is to establish working competition at the provincial level. This is because this is the level at which political control and market design are easily aligned. So it was in the US and at the individual member state level in Europe. Interstate trading has built up gradually in the US, via states voluntarily joining interstate ISOs (such as PJM and MISO). As we discussed in Chap. 4, the European single market in electricity has emerged gradually as a similar process of national markets joining with neighbours and then being finally coupled via the EUPHEMIA algorithm. A key point is that market-based inter-provincial trading can only proceed to the extent that provincial markets are themselves competitive. Once stakeholders are happy with the operation of the market within

5  Prospects for Reform of China’s Electric Power Sector 

229

individual provinces it will be possible to have a market operating across provinces. This opening up of the market to extensive provincial trading can be done gradually. It will be important to model the implications of extending trading between provinces so that the price effects can be more accurately predicted and major reallocations of welfare between the provinces and within them can be managed. Provinces with initially low prices and electricity surpluses (e.g. Inner Mongolia) need to consider how rising prices should be handled and if existing customers within the province need to be compensated from the extra revenue from electricity exports. One of the lessons from the US and from Europe is that this willingness to expand cross-jurisdictional trading is by no means guaranteed and likely to be limited by unwillingness of one of the jurisdictions involved (often the one with the lower prices) to build the necessary transmission infrastructure.

5.3.7 P  ay Attention to Mitigation of the Social Effects of Power Sector Reform A key learning from power market reform is that it does have wider external impacts which must be managed. A good example of this is the impact of power sector reform on equipment manufacturers and on the coal industry. For equipment manufacturers power sector reform can be very positive, in that it creates more competition for cost-reducing investments which meet customer needs (e.g. in increased ability to run plant flexibly in response to fluctuating market prices). However for the coal industry it may mean falling demand for coal and pressure to reduce (domestic) coal prices. This is because there will be a new emphasis on reducing input costs and competitively sourcing fuel. In the UK, the introduction of a competitive wholesale power market ended preferential coal contracts for domestic coal to supply UK power stations. It eventually led to a significant reduction in the demand for domestic coal due to a combination of its high cost, the improved thermal efficiency of the coal fleet, the switch to gas-fired generation and, most recently, exposure to significant carbon prices.

230 

M. G. Pollitt

Power sector reform in China will reduce the demand for coal, relative to a counterfactual of an unreformed power sector. To the extent to which this does lead to reductions in the demand for coal there may be a need for compensation to be given to certain coal communities affected by absolute demand reduction. Similarly if coal-fired plants are forced to close there may be local social impacts due to the effects on local employment. Reform should involve mitigation of these effects in a socially responsible way. However the mitigation should not be in the form of maintaining the output and use of coal, given its damaging effects on the local and global environment. Much attention was given to the impact of declining coal demand from the power sector in the UK as a result of power sector reform and mitigating the local social impact on coal communities involved substantial government investment. By contrast the closure of coal-fired power plants was not such a big social issue as the workers could be compensated directly from the generation companies’ own cost savings and there were virtually no compulsory redundancies due to workers’ ability to find other jobs. There will also be additional financial effects on the viability of many of the firms involved in the power industry and its supply chain. This may require the write-down of state-owned assets and transfers of funds between central and provincial governments as the policy reduces the value of generation companies, potentially threatening both their asset value and their ability to finance their existing debts and dividend payments.

5.4 F undamental Questions Raised by China’s Power Market Reform We end with some big picture issues raised by China’s power market reforms. Power sector reform has been tried in many countries and sub-­ national jurisdictions across the world, but has only been obviously successful in a few (e.g. Great Britain, Norway, Texas). Even in those few it remains controversial. The overarching question of whether China is likely, ultimately, to have a successful power market reform remains to be seen.

5  Prospects for Reform of China’s Electric Power Sector 

231

5.4.1 Is China Ready for the Full Implications of Electricity Markets? Electricity markets which work give rise to the rapid pass-through of cost rises as well as cost falls. Electricity prices are likely to become more volatile. There will be pressure to rebalance charges and reduce cross subsidies—especially towards domestic customers. There will be electricity company bankruptcies. For instance, it is difficult to see how all of its current batch of 7000+ electricity retailers can currently survive. There is a risk that customers will be mis-sold contracts or be faced with inconvenience due to the poor service or bankruptcy of their retailers. Generators, aggregators and providers of financial hedging instruments will also be subject to failure risk. A key question is whether China’s regulatory authorities will be able to rise to the regulatory challenge posed by a fully functioning set of electricity markets. In particular, will electricity markets in China survive a serious stress test, such as a sustained period of higher wholesale prices driven by high international commodity prices? Given that reform has been politically extremely controversial in the US, Australia and the UK, when there have been periods of high prices or wide area supply failures, it remains to be seen if electricity markets in China will be more or less sustainable than they have been elsewhere. The Californian electricity crisis of 2001 still casts a long shadow over half of US states, which have not opened up their electricity markets.

5.4.2 Is There an Easier Way to Deliver the Benefits of Power Market Reform Than the Route Currently Being Pursued? If China is unwilling to further restructure its electricity supply industry, would a single buyer model not have been a simpler way to deliver the same (or higher) levels of benefits in generation and retail to the Chinese economy? SGCC and CSG could simply have switched to merit order dispatch of power plants and paid generators on either a cost or a price-­ bid basis. This could have been done without the need to create

232 

M. G. Pollitt

thousands of new retailers across China and with very little requirement for increased sophistication of the existing regulatory regime. Given that most of the trading in the power market is currently done via long-term bilateral contracting (rather than via auctions involving retailers), the amount of trading that requires retailers at all remains small. Essentially SGCC and CSG could have been tasked with massively reducing purchased generation costs and likely delivered this very efficiently and with little risk. Merit order dispatch remains the largest potential benefit of power market reform, at least in the near term. The further introduction of incentive regulation on the network companies could also have been done independently of the reform of power purchasing. This would have simply involved incentivising SGCC and CSG to control their costs (and increase their quality of service) by setting a price control formula on their total revenue less generation costs.

5.4.3 C  an Reform Be Sustained and Completed in China Given Its Institutional Set-Up? Power market reform involves the undermining of the vested interests of the power system. Fundamentally, power sector reform shifts stakeholder power away from incumbent networks and generators towards regulators and retailers who represent consumer rather than producer interests. It also reduces the power of the workers in the power system and in electricity supply chain companies. Indeed, progressing power sector reform in the US and Europe depended on not listening to the voices of incumbent companies who said that power market reform would be bad for their companies (and hence the whole economy). Power sector reform required strong and sustained political support in the face of attempts to derail it. Interestingly reliance on competitive energy markets is not a recipe for a ‘happy’ system: it is a recipe for controversy. International polls of citizens consistently show that they think that states with more energy market intervention and less reliance on energy markets have a more trusted energy industry, even when in objective terms they have ‘worse’ energy policies resulting in high overall costs, low quality of service and poor

5  Prospects for Reform of China’s Electric Power Sector 

233

environmental outcomes.12 This is more or less in line with less trust in government being lower in countries with more competitive energy markets.13 China’s power market reforms have been derailed before (in 2007) and quietly dropped. They remain vulnerable to changes of political sentiment, rising market prices of energy and the lobby power of incumbents. Chinese policy-makers need to stand ready to make a ‘there is no alternative’ (TINA) argument in support of sustaining power market reform. There is an open question as to whether effective power markets required advanced forms of democratic legitimacy. Successful power market reform has not occurred outside of sophisticated democracies (though it was begun in Chile prior to the reintroduction of democracy). High power market prices have been politically sensitive in other jurisdictions. The Chinese authorities currently seem willing to switch off the day-­ ahead market in advance if there is any possibility of it leading to high prices at particularly politically sensitive times (such as public holidays); this negates the fundamental role of short-term market prices as the basis of the price formation of longer-run prices.

5.4.4 C  an China Have a Successful Power Market Reform Without Widespread Private Ownership of the Sector? Power markets and incentive regulation work best where private companies pursue the profit opportunities provided by market signals and incentive regulation. The reforms in the US and the UK depend on private ownership for their success, especially when it comes to retail competition. It is true that public ownership remains significant in Australia, New Zealand and Norway and these countries also have competitive power markets. But there is a lot of competition between public and private  In 2018, according to the international polling firm Edelman, general population trust in the energy sector was 88% in China, against 39% in Australia, 43% in the UK, 63% in the US (see Edelman 2018). 13  In 2018, according to the international polling firm Edelman, general population trust in government was 84% in China, against 35% in Australia, 36% in the UK, 33% in the US (see Edelman 2018). 12

234 

M. G. Pollitt

firms in these jurisdictions. These countries are also small relative to a typical Chinese province and have a tradition of profit-­oriented public ownership (corporatisation). The global track record of large state-owned firms in pursuing profits is much less good, as we discussed in Chap. 2.

5.4.5 Is China Willing to Break Up SGCC and CSG in Ways That Will Promote Power Market Development? SGCC and CSG are still much too dominant within the domestic electricity supply sector in China. At the very least they need to be effectively legally unbundled into a network business, a system operation business and retail business. However their retail businesses need to be broken up and fully separated from the rest if there is to be genuine retail competition. Otherwise the ownership structure in China will diverge from all other deregulated power markets in the world. This would seem to be a problem for Chinese power equipment manufacturers trying to understand developments with distributed generation, prosumers and cross-­ platform energy market competition in the rest of the world. In the rest of the world we might well see former oil companies retailing electricity, direct competition between utilities and their customers for electricity supply and major non-energy retailers (such as Amazon, Google and supermarket chains) enter the retail electricity market. The equivalent (e.g. Sinopec, PetroChina, Alibaba, Baidu, Sun Art entering the electricity market) will not happen in China without the restructuring of SGCC and CSG.

5.4.6 H  ow Will China Combine Power Sector Reform with Decarbonisation? We saw that 40% of the industrial price of electricity in the UK was determined by the government and that this was split roughly between carbon pricing and renewables support. China needs to incentivise and pay for the decarbonisation of its power sector. A carbon price of $25 per

5  Prospects for Reform of China’s Electric Power Sector 

235

tonne of CO2 (in line with the EU) could increase the 2014 industrial price of electricity by over 20%.14 China’s industrial electricity price in 2018 included government charges for debts on network extension investments, the Three Gorges Dam, an urban utility surcharge and a renewable development surcharge.15 Substantial decarbonisation will add further to the costs of electricity and offset any price gains due to efficiency from power sector reform. However continuing to support renewables and further developing the role of carbon markets in China is essential to achieving China’s environmental goals.

5.4.7 W  hat Will Reform Reveal About Chinese Electricity Consumers? Finally, if reform changes the orientation of the electricity system from being about what electricity producers want to what electricity consumers want, what will this reveal about consumer preferences? There is an assumption at the heart of any unreformed electricity system that the electricity industry knows what is best for its customers. Power market reform fundamentally changes this perspective. Consumers decide on what terms they want to purchase power and express preferences as to what they are willing to pay for. Regulators acting in the interests of consuming companies and citizens question the need for investments and their efficacy. They focus on output measures that consumers and citizens value, such as price, environment performance and quality of service, over multiple dimensions such as speed of response to enquiries. Power market reform has revealed and is revealing many things in the jurisdictions that have undertaken it, about what it is that consumers want.16 Two major things that consumers appear to want are to be treated fairly and to not be exposed to unexpected rises in their bills. Relatively few customers want to be exposed to nodal prices (indeed few generators  If the marginal fuel in the Chinese power system is coal and there is roughly 1 kg of CO2 emissions per kWh delivered, this gives an increased cost of US $0.025 per kWh (the industrial price in 2014 was US $0.1068 per kWh). 15  See Alva and Li (2018, p. 51). 16  See, for example, Sioshansi (2019) for a recent perspective on this. 14

236 

M. G. Pollitt

want to be exposed to these). However differing numbers of customers are willing to sign more sophisticated power contracts, but this often depends on the size of their electricity bill. Customers are sometimes willing to pay for things like undergrounding or offshoring of electricity assets. However often what they might say about willingness to pay for individual investments is at odds with their unwillingness to see a rise in their overall bill. What increased attention to customer needs will reveal about willingness to pay in China for electricity is something that power market reform can and should reveal.

5.5 K  ey Closing on Messages for Chinese Electricity Stakeholders on How to Approach Power Sector Reform We conclude with some very high-level positive messages for the different stakeholders within the electricity supply industry in China, drawing on international experience. These messages emphasise successful strategies followed by such stakeholders in other reformed electricity markets.

5.5.1 Policy-makers The overall social welfare benefits of power sector reform are positive but small. The major benefit is the reduction of the political risk arising within the electricity sector of the state taking responsibility for the management of the power sector. The outsourcing of the power sector to ‘the market’ reduces the role of the state to making sure that regulatory institutions around the market function effectively. The difficulty of making strategic decisions within the system is only increased by the need to decarbonise electricity production. The European Commission has been successful at promoting power market reform across Europe, while policy-­makers in Texas have created arguably the most competitive state-­ level market in the US.  Both have done this while seeing significant increases in low carbon generation.

5  Prospects for Reform of China’s Electric Power Sector 

237

5.5.2 Regulators The regulation of the electricity system essentially comes down to monitoring whether the power system is working in the interests of the electricity consumer. This involves monitoring and promoting competition in wholesale and retail markets and the implementation of effective incentive regulation of power networks. Concentrating on these two elements is the job of the economic regulator for the power system. Good examples of regulatory agencies that do this are Ofgem in the UK and the AER in Australia.

5.5.3 Generators Power markets focus attention on efficient operation of power plants and good investment decision making around future investment. Generation companies that focus on being good at power plant operation and on effectively modelling future market trends, appropriately hedging market risks and managing new power plant costs, can be profitable and beat the industry average performance. Good examples of generation companies that do this are Engie from Belgium and Duke Power in the US.

5.5.4 Retailers Competitive power markets focus on giving consumers what they value. Retailers should be interested in fair competition in retail markets and on building a value proposition which makes sense to their customers. Strong retail companies that focus on their customer in terms of delivering value for money with appropriate levels of customer service can be profitable. Good examples of retail energy companies that have been successful in the competitive era are Centrica in the UK and NRG in the US.

238 

M. G. Pollitt

5.5.5 Grid Companies The core network monopoly will survive the introduction of competitive markets for wholesale and retail power. It is in the interests of grid companies to get out of the retail business and focus being a network availability provider. Grid companies should focus on cost reduction and provision of high-quality service and be efficiently divided into transmission, distribution and system operation. They should aspire to be trusted third parties by all other stakeholders within the power system. Good examples include UK Power Networks and the UK’s National Grid.

References AER. (2018). State of the energy market 2018. Melbourne: Australian Energy Regulator. Alva, H. A. C., & Li, X. (2018). Power sector reform in China: An international perspective. Paris: IEA Publications. Edelman. (2018). 2018 Edelman trust barometer: Attitudes toward energy in a polarized world. Available at: https://www.slideshare.net/EdelmanInsights/ 2018-edelman-trust-barometer-attitudes-toward-energy-in-a-polarizedworld EIA. (2019). Electric power monthly with data for June 2019. Washington, DC: US Energy Information Agency. FERC. (2019). State of the markets report 2018. Washington, DC: Federal Energy Regulatory Commission. Monitoring Analytics LLC. (2019a). State of the market report for PJM, Volume I: Introduction. Eagleville, PA: Monitoring Analytics LLC. Monitoring Analytics LLC. (2019b). State of the market report for PJM, Volume II: Detailed analysis. Eagleville, PA: Monitoring Analytics LLC. NBS. (2018). China electric power yearbook 2018. Beijing: China Electric Power Press. NDRC. (2019). Opinions on deepening the electricity spot market construction in pilot areas. Available at: http://www.ndrc.gov.cn/gzdt/201908/ t20190807_943964.html Ofgem. (2018). State of the energy market 2018. London: Ofgem.

5  Prospects for Reform of China’s Electric Power Sector 

239

Pollitt, M. (2019). The single market in electricity: An economic assessment. Review of Industrial Organization, 55(1), 89–109. Rawski, T.  G. (2019). Growth, upgrading, and excess cost in China’s electric power sector. In L.  Brandt & T.  G. Rawski (Eds.), Policy, regulation, and innovation in China’s electricity and telecom industries (pp.  304–372). Cambridge: Cambridge University Press. Sioshansi, F. P. (2016). What is the future of the electric power sector? In F. P. Sioshansi (Ed.), Future of utilities – Utilities of the future: How technological innovations in distributed resources will reshape the electric power sector (pp. 3–23). London: Elsevier. Sioshansi, F. (Ed.). (2019). Consumer, prosumer, prosumager: How service innovations will disrupt the utility business model. London: Academic Press.

Index1

A

ACER, see Agency for the Cooperation of Energy Regulators Adib, P., 140n91 AER, 237 Agency for the Cooperation of Energy Regulators (ACER), 59n69, 115 Aghion, P., 80n116 Air pollution, 78, 79, 89, 108 Alibaba, 234 Allocated hours, 47 Alva, H.A.C., 7n6, 57n66, 218n2, 218n3, 224n10, 235n15 Amazon, 234 An, Bo, 63n77

Anaya, K.L., 23n2, 24, 26n6, 41n41, 46n48, 71n93, 75n99, 76n104, 200n58, 203n62 Ancillary services, 114, 140, 162, 169, 193, 198–200, 207 Chinese context, 46–47 general reform experience, 46 theoretical significance, 44–45 Andrews-Speed, P., 106n8, 114, 114n26 ANEEL, 61 Anti-monopoly Act of 2018, 132 Anti-monopoly Law of 2015, 65 Anti-trust/anti-monopoly, 40, 65, 66, 208 APX, 115n28 APX Power UK, 160

 Note: Page numbers followed by ‘n’ refer to notes.

1

© The Author(s) 2020 M. G. Pollitt, Reforming the Chinese Electricity Supply Sector, https://doi.org/10.1007/978-3-030-39462-2

241

242 Index

Archer, C.L., 53n60 Argentina privatisation and monopolies, 41 Auction, 53, 59, 70, 71, 73, 74, 172, 198, 200, 205, 232 Australasia, 2, 216, 218 wholesale energy market, generation and demand in, 227 Australia Australian Energy Regulator, 225 full implications of electric power, 231 independent system operators, 198 power sector reform without private ownership, 233 regulatory reporting, improvement of, 225, 226 Ausubel, L., 136n82 B

Baidu, 234 Balancing and Settlement Code (BSC), 159 Balancing market/mechanism, 44–47, 114, 121, 161–163, 166, 190, 192, 194, 198 Balancing mechanism units (BMUs), 163, 164, 190–193 Balancing service use of system charge (BSUoS), 190 Beijing, 129, 218 system operators, 36 BEIS, see Department for Business, Energy and Industrial Strategy Belgium, 168

Bergman, L., 28n12, 53n61 Bessant-Jones, J.E., 27n7, 61n76 Bessembinder, H., 114n27 Bialek, J., 76n103 Birmingham transmission charges, 187 Bishop, M., 41n40 Blackouts, 46, 48, 80 Blanchard, O.J., 80n116 BMUs, see Balancing mechanism units Bohn, R.E., 52n56, 188n50, 207n72 BP, 25n4 Brazil privatisation and monopolies, 41 Bristol Energy, 160n12 British Electricity Trading & Transmission Arrangements (BETTA), 197, 200 British Energy, 121, 167 British Gas, 120, 170, 172 privatisation of, 41 BritNed, 187 BSC, see Balancing and Settlement Code BSUoS, see Balancing service use of system charge Bui, T., 106n8 BYD, 124 C

California, 25, 26, 107 electricity crisis of 2000–01, 29, 162, 231 Renewable Auction Mechanism, 71, 74 Carbon dioxide (CO2), 38, 69, 76–78, 115, 164, 206

 Index 

Carbon Emissions Reduction Target (CERT), 170 Carbon markets, 25, 78, 108, 164 Carbon price floor (CPF), 206 Carbon price support (CPS), 204, 206 Carbon pricing, 69, 108, 154, 164, 167, 204, 206 switching, 90–92 Carbon reduction commitment (CRC), 204, 206 Carbon trading, 108 CC, see Competition Commission CCL, see Climate change levy CEEC, see China Energy Engineering Corporation CEGB, see Central Electricity Generating Board Central Electricity Generating Board (CEGB), 41, 42, 42n45, 113, 120, 156, 199 Centrica, 237 CER, see Certified Emission Reductions CERT, see Carbon Emissions Reduction Target Certified Emission Reductions (CER), 107 CfD, see Contract for difference Chao, H.P., 53n59 Chau, J, 103n4 Chen, H., 38n36, 79n115 Chen, Q., 71n94 Chen, R., 129n67 Chen, X., 15n18 Cheng, B., 107n11, 107n12, 108 Cheng, C., 34n24, 138n85 Cheng, W., 103n2 Chick, M., 113n23

243

Chile privatisation and monopolies, 41 transmission capacity, 52 vertical separation and horizontal restructuring, 28 China Datang Corporation, 10 China Electricity Council, 4n3, 12n13 China Energy Engineering Corporation (CEEC), 12, 12n12 China Energy Engineering Group, 125 China Energy Investment, 10 China Energy Storage Alliance (CNESA), 41n42 China5e Research Centre, 108n16 China General Nuclear (CGN), 11 China Guodian, 10 China Huadian Corporation, 10 China Huaneng Corporation, 10 China Huaneng Group, 132n78 China Light and Power Group (CLP), 128, 128n63 China National Development and Reform Commission, 81n118, 108n16 China National Energy Agency, 65n80, 78n110 China Resources Power (CRP), 10, 124 China Southern Grid (CSG), 10–12, 14, 30–32, 37, 42, 57, 67, 84, 87, 103, 104, 116–118, 124, 126, 128, 130–132, 137, 138, 172, 208, 215, 222, 222n9, 227, 231, 232, 234

244 Index

China Southern Power Grid (CSPG), 128n64 China State Council, 1, 1n1, 25n5 Chinese electric power benefits of reform, 231–232 full implications of, 231 sustained and completed reform, 232–233 in 2015, scale and scope of, 4–9 Chinese power sector history of reform, 13–16 reform timeline for, 14 structure and organization of, 10–12 2015 reform, motivation for, 16–19 CHPs, see Combined heat and power plants Chuanlong, Xu, 108n17 Ciarreta, A., 70n89 Climate change levy (CCL), 204, 205 CLP, see China Light and Power China Energy Storage Alliance (CNESA), 41n42 Coal, 38, 39, 77–79, 81, 85–88, 90, 108, 110, 115, 117, 128, 129, 156, 167, 168, 208 Cohen, Reinhold, 107n10 Combined Cycle Gas Turbine (CCGT), 189 Combined heat and power plants (CHPs), 39 Competition Act, 174 Competition and Markets Authority (CMA, UK), 128, 128n60, 160, 161, 161n13, 162n15, 165, 171, 174, 188n49

Competition Commission (CC), 176 Competition policy, 225 Competitive procurement of wholesale power, for regulated final customer groups Chinese context, 59–60 general reform experience, 59 theoretical significance, 58 Competitive procurement process, for low carbon generation Chinese experience, 71–73 general reform experience, 70–71 theoretical significance, 69–70 Competitive segment charges Chinese context, 57–58 general reform experience, 56 theoretical significance, 55 Competitive segments, monitoring, 60–68 Connection charging, 184 Contract for difference (CfD), 70, 71, 164, 204, 205 CORESO, 164 CPF, see Carbon price floor CPS, see Carbon price support Cramton, P., 136n82 CRC, see Carbon reduction commitment Crossley, D., 49n53 CRP, see China Resources Power CSG, see China Southern Grid CSPG, see China Southern Power Grid Cunningham, E.A., 29n17 Currier, K.M., 69n87, 70n88 Curtailment, 38, 54, 76, 86, 88

 Index  D

Dai, H., 140n93 Datang, 222n8 Decarbonisation and power sector reform, combining, 234–235 DECC, 138n87, 156n5, 156n6, 205n69 DECC, see Department for Business, Energy & Industrial Strategy Deepening Reform of the Power Sector, 1 Demand Side Balancing Reserve (DSBR), 200 Department for Business, Energy and Industrial Strategy (BEIS), 161 Department for Business, Energy & Industrial Strategy (DECC), 71n91 Deregulation, 64, 81, 234 Development, 12, 46, 66, 69, 87, 106, 107, 114, 117, 167, 171, 182, 189, 203, 208, 215–224, 234 Digest of United Kingdom Energy Statistics 2018, 156n3 Discount, 108, 116, 117, 124, 125, 130, 131, 133, 135, 136 Dispatch, 34–39, 47, 53, 85, 87–89, 105, 113, 114, 137, 139, 161, 163, 164, 168, 169, 188, 189, 197, 207 wholesale markets effect on Guangdong, 128–130 international context, 126–128

245

Distribution, 4, 10, 11, 15, 25, 25, 27, 27, 27, 27, 28, 28, 29, 31, 31, 31, 31, 31, 31n21, 31, 31, 32, 32, 32, 32, 32, 41, 41, 41, 42, 51, 53, 54, 54, 55, 55, 55, 55, 55, 55, 56, 56, 56, 56, 57, 57, 57, 57, 58, 58, 58, 58, 58, 59, 61, 62, 67, 67, 71, 104, 104, 117, 117, 118, 118, 122, 130, 130, 130, 138, 156, 157, 157, 157, 159, 159, 159, 163, 164, 164, 164, 171, 173, 173, 175, 175, 179, 179, 181, 182, 189, 191, 191, 191, 191, 205, 205, 205, 208, 208 charges, 3, 170, 181, 182, 201–203, 205, 207 Distribution network operator (DNO), 182 Distribution system operators (DSOs), 199 Distribution use of system charges (DNUoS), 202 DNO, see Distribution network operator DNUoS, see Distribution use of system charges Document No. 5, 14 Document No. 9, 1, 9, 16, 17, 31, 67, 86, 87, 104, 140, 140n92, 153 Domah, P.D., 41n41, 201n59 Drummond, P., 204n65 DSBR, see Demand Side Balancing Reserve

246 Index

DSOs, see Distribution system operators Duke Power, 237 Dupuy, M., 78n109 E

ECO, see Energy Company Obligation Economic regulation competitive procurement of wholesale power, for regulated final customer groups, 58–60 competitive segment charges, 55–58 regulated network charges, 55–58 regulatory agencies, for monopoly network charges regulation and monitoring competitive segments, 60–68 Economics and Information Commission (EIC), 108, 131 EdF, 10, 11, 41, 113, 159 EES, see Electrical energy storage Effective allocation, theoretical significance of, 49–52 EFR, see Enhanced frequency response EIC, see Economics and Information Commission ElecLink, 188 Electrical energy storage (EES), 199 Electricity Act of 1996, 140, 140n92, 174 Electricity consumers, need of, 235–236

Electricity market pilot projects (2016), 83–84 Electricity Northwest, 159 Electricity production mix, 221 Electricity trading, 36n32, 83, 84 Elexon, 159 Ellerman, A.D., 25n3, 77n106 Emissions, 7, 8n6, 23–25, 38, 69–79, 107, 115, 164, 206, 216 Energy Company Obligation (ECO), 170 Energy Contract Volume Notifications (EVCN), 163 Energy efficiency, 124, 125, 156, 170, 171, 174, 204–208 Engie, 237 England, see United Kingdom (UK) Enhanced frequency response (EFR), 198 Environmental externalities, pricing of Chinese experience, 78–79 general reform experience, 77–78 theoretical significance, 76–77 Environmental levies and taxes, 204–207 E.ON, 159 ERCOT, 114 EU, see European Union EU ETS, see European Union Emissions Trading Scheme EUPHEMIA algorithm, 115, 169, 228 Europe, 2, 216, 218 exposure to wholesale market prices, 128

 Index 

interprovincial market, creation of, 228, 229 new energy market players, 121 policy-makers, 236 power markets, 115 progressing power sector reform in, 232 retail margins, 171 transmission charges, 189 wholesale energy market, generation and demand in, 227 European Commission, 236 European Union (EU), 25, 115, 227, 227n11, 235 carbon dioxide, permit scheme for, 78 competitive procurement of wholesale power, for regulated final customer groups, 59 regulated network and competitive segment charges, 56, 57 regulatory agencies, for monopoly network charges regulation and monitoring competitive segments, 61 system operators, 36 transmission capacity, 52 transmission charges, 187 European Union Emissions Trading Scheme (EU ETS), 25, 78, 108, 204, 206, 217 Evans, J., 166n26 EVCN, see Energy Contract Volume Notifications

247

F

Federal Electricity Regulatory Commission (FERC), 61, 66, 139n90 Federal Energy Regulatory Commission (FERC), 225, 227 Feed-in tariff (FIT), 69, 70, 73, 164, 204 Feng, Y.-C., 59n73, 138n85 FERC, see Federal Electricity Regulatory Commission; Federal Energy Regulatory Commission FFR, see Firm frequency response Final Physical Notification (FPN), 163, 192, 193 Finance Ministry (UK), 154 Financial transmission rights (FTRs), 53, 189 Firm fast response, 198 Firm frequency response (FFR), 198 FIT, see Feed-in tariff Five Year Plan (FYP) 12th, 64 13th, 8 Florio, M., 41n40 Fossil fuels, 25, 29, 32–34, 38, 57, 60, 69, 70, 73, 74, 76, 78, 81, 86, 88, 133, 155, 164, 166, 206, 207 Fowlie, M., 77n107 FPN, see Final Physical Notification France, 168 Frequency response, 44, 162, 196, 198 FTRs, see Financial transmission rights

248 Index

Fujian, 219, 221 characteristics of, 220 electricity balances in, 221 electricity production mix in, 221 progress with power market implementation, 222 Future research, suggestions for, 90 FYP, see Five Year Plan G

Gansu, 219 characteristics of, 220 electricity balances in, 221 electricity production mix in, 221 progress with power market implementation, 222 Gas Act, 174 GdF-Suez, 120 GDP, see Gross domestic product GE, 15 GEDI, see Guangdong Electric Power Design Institute Co. Ltd. Generation mix, 11 Generators, 11, 119, 120, 154–155, 157, 159–168, 174, 184, 186, 189–191, 193, 196, 198, 200–202, 204, 237 Germany, 25, 26, 85 environmental levies and taxes, 206 spot markets and ancillary services, 46n48 vertical separation and horizontal restructuring, 28 GHG, see Greenhouse gas Google, 234

Governance, 106, 107, 160 Great Britain, see United Kingdom (UK) Green, R., 166n26 Greenhouse gas (GHG), 79 Greve, T., 198 Grid companies, 36, 57, 172, 238 Gross domestic product (GDP), 72, 86, 103, 106, 107, 140 Grubb, M., 69n87, 204n65 Guangdong, 3, 10, 216–219, 223, 224 background of, 106–109 characteristics of, 220 electricity balances in, 221 electricity production mix in, 221 electricity sector, size of, 109–113 industrial electricity price, determination of, 155 infra-marginal bids and auction changes, 141 learning from pilot markets, promoting, 226 network charges, 182 new energy market players, 122–126 overall impressions of reform, 130–139 power market, 116–119 power market reform, 103–145 progress with power market implementation, 222 recommendations for furthering reform, 139–140 regulated network and competitive segment charges, 57 retail margins, 172

 Index 

spot market, 163 system operators, 36 transmission capacity, 53 wholesale markets effect on dispatch, 128–130 wholesale prices, 168 Guangdong Development and Reform Commission (DRC), 131 Guangdong Electric Power Design Institute Co. Ltd. (GEDI), 125, 126 Guangzhou, 106, 107, 128, 129, 218, 219 Guangzhou Emissions Exchange (GZX), 107 Guangzhou Power Exchange Center, 109n19 Guodian, 222n8 GZX, see Guangzhou Emissions Exchange H

Hebei system operators, 36 Heller, T.C., 41n42 Helm, D., 205n66 Henney, A., 139n90, 140n94, 156n7, 165, 170n37 Hermann, C., 122n49 High Voltage Direct Current (HVDC), 53 Hirst, D., 167n29 HM Treasury (UK), 154 Ho, M.S., 129n65 Hogan, W.W., 49n55, 188n50 Horizontal restructuring, 26–29

249

general reform experience, 28–29 theoretical significance, 27–28 Hove, A., 75n101 Huadian, 222n8 Huaneng, 222n8 Hubei, 10 Hurlbut, D., 114n24 Hydro-Benefit Replacement Scheme, 205 Hydro-Benefit Scheme, 204, 205 Hydropower, 36, 128, 130, 138, 205 I

Iberdrola, 159 IEA, see International Energy Agency Incentive regulation, 42, 56, 58, 62, 67, 87, 88, 122, 130, 196, 201, 203, 208 INDC, see Intended Nationally Determined Contribution Independent Energy, 121 Independent system operators (ISOs), 36, 53, 198, 227, 228 Industrial electricity prices, determination of, 153–208 environmental levies and taxes, 204–207 key actors, 156–161 method, 154–155 network charges, 173–182 retail margins, 169–173 system balancing charges, 190–200 distribution charges, 201–203 transmission charges, 182–190 wholesale prices, 161–169

250 Index

Industrial policy, 208 Inner Mongolia, 219, 221 characteristics of, 220 electricity balances in, 221 electricity production mix in, 221 interprovincial market, creation of, 229 progress with power market implementation, 222 Inner Mongolia Power Corporation, 12 Institutions, 65, 87, 107, 156, 225, 232–233, 236 Integrated Transmission Planning & Regulation Project (Ofgem), 197 Intended Nationally Determined Contribution (INDC), 79 International Energy Agency (IEA), 7n4, 7–8n6, 59n70, 103n3 Interprovincial market, creation of, 228–229 Investment, 3, 4, 6, 12, 27, 34, 38, 40–42, 49, 58, 62, 69, 78, 82, 85, 86, 88–91, 104, 115, 125, 127, 139, 159, 168, 174–177, 181, 183, 188, 189, 196, 200, 206–208, 226, 229, 230, 235s7 Ireland system operators, 35 Irish Electricity Supply Board, 188 ISOs, see independent system operators Italy competitive procurement of wholesale power, for regulated final customer groups, 59

J

Jackson, R.B., 79n114 Jamasb, T., 28n9, 56n64, 58n67, 61n76, 122n47, 165n22, 177n43 Jiangsu demand response, potential for, 51 pilot markets, 226 power sector reform, recent developments in, 222–223 wholesale electricity markets, demand-side participation in, 49 Joskow, P.L., 23–25, 23n1, 28n10, 55 K

Kahrl, F., 30n18, 37n33, 71n94, 81n117 Kim, T.-H., 42n45 Klemperer, Paul, 136n82 Krishna, V., 116n31 Kuang, J., 140n93 Küfeoğlu, S., 203n61 L

Latin America system operators, 35 Lei, N., 49n52 Lemmon, M.L., 114n27 Levies, 169, 204–207 Li, M., 116n32 Li, X., 7n6, 57n66, 76n102, 218n2, 218n3, 224n10, 235n15 Li, Y., 57n65, 132, 132n77 Lin, J., 114 Lin, K.C., 15, 15n17

 Index 

Littlechild, S.C., 58n68, 59n71, 175 Liu, S., 34n24, 138n85 Liu, Z., 71n94 Lizi, Zhang, 108n17 LMPs, see Locational marginal prices Locational marginal prices (LMPs), 51–53, 188, 189 Lohmann, P., 122n48 London transmission charges, 187 London Stock Exchange, 159 Low carbon, 23–26, 89, 164, 197, 206 generation, competitive procurement process for, 69–73 Low emissions, power market reform in, 24 Low emission technologies, efficient promotion of competitive procurement process, 69–73 environmental externalities, pricing of, 76–79 renewables, cost reflective access terms for, 74–76 M

Ma, J., 59n72 Mansur, E.T., 35n26, 115, 115n30, 128n61, 169n35 Marino, A., 122n46 Market coupling, 169 Market reform, 1–19, 23, 24, 29, 31, 52, 57, 80–87, 89, 103–145, 153–208 Market restructuring and ownership changes Chinese context, 29–34 privatisation and monopolies, 40–44

251

system operators, 34–40 vertical separation and horizontal restructuring, 26–29 Mathews, J.A., 81n117 Menezes, F.M., 81n118 Merit order, 34, 36, 37, 39, 87, 113, 137 Midcontinent Independent System Operator (MISO), 114, 228 Ming, Z., 47n49, 47n50 Mingtao, Y.A.O., 47n50 Ministry of Communications, 64 Ministry of Electric Power, 10 Ministry of Environmental Protection, 64 Mitigation of social effects, 229–230 MMC, see Monopolies and Mergers Commission Mo, K., 75n101 Monopolies and Mergers Commission (MMC), 176 Monopoly network charges, regulation of, 60–68 Monopoly state ownership Chinese context, 41–44 general reform experience, 41 theoretical significance, 40–41 Mota, R.L., 41n41 Multi Regional Coupling, 169 N

N2EX, 160 National Bureau of Statistics (NBS), 7n5 National Development and Reform Commission (NDRC), 38n35, 46, 49n53, 72, 78, 82, 118, 119, 130, 131, 218 Pricing Departments, 63

252 Index

National Grid (UK), 35, 157, 159, 160, 164, 164n19, 186–188, 186n47, 186n48, 196–198, 203n62, 238 National Grid Electricity Transmission (NGET), 182–184, 196, 197 controllable operating costs, 183 loss of supply incidents, 184 2008–09 reliability incentive scheme, 187 Natural resources, 208 NBS, see National Bureau of Statistics NDRC, see National Development and Reform Commission NEA (the Electricity Regulatory Commission), 63, 67, 81n118, 131, 208, 218, 226 NETA, see New Electricity Trading arrangements Net imbalance volume (NIV), 194, 194n54 Net present value (NPV), 127 Netherlands, 168 privatisation and monopolies, 41 retail margins, 171 Network charges, 54–58, 118, 121, 131, 132, 160, 169, 173–182 Network Rail, 42 Neuhoff, K., 69n87 Newbery, D.M., 25n3, 29n13, 35n27, 40n39, 45n47, 126n59, 156n7, 165n22, 196n55, 207n71 New Electricity Trading Arrangements (NETA), 159, 166, 189, 190, 197, 200

New energy market players in Guangdong, 122–126 international context, 119–122 New normal, 4, 218 New York, 26 competitive procurement process, for low carbon generation, 70 spot markets and ancillary services, 46n48 New Zealand, 171n38 market restructuring and ownership changes, 31n21 power sector reform without private ownership, 233 retail margins, 171, 171n39 Ng, E., 128, 128n62 NGET, see National Grid Electricity Transmission Nicolli, F., 156n4 NIE, see Northern Ireland Electricity Nillesen, P.H.L., 31n21, 171n38 Nitrous oxide (NOX), 76, 78, 79 NIV, see Net imbalance volume North America, 7n5 exposure to wholesale market prices, 128 new energy market players, 121 Northern Ireland, 154n1 transition mechanisms, 81 Northern Ireland Electricity (NIE), 159 Northern Powergrid, 159 Norway, 2 power sector reform without private ownership, 233 NOX, see Nitrous oxide NRG, 237 Nuclear power, 37, 71, 72, 204

 Index  O

O’Donnell, A.J., 34n25 Offer, 62, 165, 166n27 Ofgem, 61, 62, 120n41, 121n42, 154n1, 160, 165, 165n23, 168n32, 170, 173–175, 177, 177n43, 177n44, 179, 181, 181n45, 181n46, 200n57, 204, 225, 237 Integrated Transmission Planning & Regulation Project, 197 menu regulation scheme, 177 Onaiwu, E., 163n17 ‘One-Belt One-Road’ policy, 125 Opinions on Deepening the Electricity Spot Market Construction in Pilot Areas, 218 P

PA Consulting Group, 160n10 PAR, see Price Average Reference Volume Paredes, R., 126n58 Peck, S., 53n59 Peru privatisation and monopolies, 41 PetroChina, 234 Pike, Lili, 108n14 Pilot markets learning from, promoting, 226–227 Pilot provinces, 220–222 Pingkuo, L., 37n34 PJM, 35, 55, 105, 114, 119, 121, 128, 137, 161, 163, 223, 225, 228 State of the Market Report for PJM, 225

253

Policy-makers, 236 Policy priorities, 85–88 Pollitt, M.G., 23n2, 24, 26n6, 28n9, 28n11, 31n21, 35n28, 36n29, 41n41, 42n44, 46n48, 56n64, 58n67, 61n76, 71n93, 75n99, 76n103, 76n104, 122n47, 126n58, 126n59, 156n7, 165n22, 169n34, 171n38, 175, 177n43, 189n51, 200n58, 201n59, 203n60–62, 227n11 Pollution, 77, 78, 89 Pond, R., 122n49 Pool Price Enquiry December 1991, 165 Pool price statement July 1993, 165 Power exchange centre, 104 Power markets in Guangdong, 116–119 international context, 113–116 reform in low emissions, 24 Price Average Reference Volume (PAR), 194 Price control, 58, 81, 170, 174–177, 179, 182, 232 Privatisation, 29, 56, 62, 85, 156n7, 167, 175, 183, 196, 197, 201, 201n59 Chinese context, 41–44 general reform experience, 41 theoretical significance, 40–41 whole or part, 228 Public Utility Commissions (PUCs), 61, 67 PUCs, see Public Utility Commissions

254 Index Q

Quality of service, 31, 61, 110, 112, 173, 179, 180, 184, 201, 202 R

RAB, see Regulatory asset base Rahimi, A.F., 28n8 RAM, see Renewable Auction Mechanism RAP, see Regulatory Assistance Project Rawski, T.G., 218n1 RECLAIM scheme, 77, 79 Reform, 1–19, 23–90, 103–145, 153–208, 215–238 Regulated network charges Chinese context, 57–58 general reform experience, 56 theoretical significance, 55 Regulated price, 33, 59, 116, 125, 126, 133, 135 Regulated third-party access, theoretical significance of, 49–52 Regulators, 32, 40, 55, 59–64, 66, 67, 89, 90, 132n76, 155, 160, 161, 165, 168, 170, 173–177, 181, 183, 187, 188, 196, 201, 208, 217, 232, 235, 237 Regulatory agencies, for monopoly network charges regulation and monitoring competitive segments Chinese context, 63–68 general reform experience, 62–63 theoretical significance, 60–62

Regulatory asset base (RAB), 176 Regulatory Assistance Project (RAP), 117n33 Regulatory capacity, improvement of, 225 Regulatory reporting, improvement of, 225–226 REMIT legislation, 165 Renewable Auction Mechanism (RAM), 71, 74 Renewable energy, 38n35, 122–124 non-hydro, 72 Renewable Energy Law (2010), 38, 38n35 Renewable obligation certificates (ROCs), 204 Renewables, cost reflective access terms for Chinese experience, 75–76 general reform experience, 74–75 theoretical significance, 74 Report on Constrained-On Plant Oct 1992, 165 Residual Supply Index, 28, 29 Retail companies, 41, 120, 121, 156, 157, 170, 199 Retailers, 25, 32, 35, 58, 80–82, 104, 108, 109, 115–126, 130, 132, 133, 135, 136, 139, 141, 143, 145, 154, 157, 159, 161–164, 166, 168–173, 184, 198, 204, 208, 216, 218, 219, 221, 222, 222n8, 225, 227, 231, 232, 234, 237 Retail margins, 169–173 RIIO, 181–182 ROCs, see Renewable obligation certificates

 Index 

Royal Mail, 42 RPI-X, 175, 181, 182 Russia, 7n5 RWE, 159 S

SAMR, see State Administration for Market Regulation SASAC, see State-owned Assets Supervision and Administration Commission SBR, see Supplemental Balancing Reserve Schmalensee, R., 55 Scotland system operators, 35 transmission capacity, 52n57 wholesale prices, 166, 167 Scottish and Southern Energy (SSE) Power Distribution, 159 SEGBA, 41 SESS, see Shenzhen Energy Sales and Services Company SGCC, see State Grid Corporation of China Sha, F., 79n114 Shandong, 219 characteristics of, 220 electricity balances in, 221 electricity production mix in, 221 power sector reform, recent developments in, 219 progress with power market implementation, 222 Shanghai demand response, potential for, 51

255

SOX emissions, 79 wholesale electricity markets, demand-side participation in, 49 Shanxi Anti-monopoly Act, 132n76 electricity balances in, 221 electricity production mix in, 221 progress with power market implementation, 219, 222 Sheffrin, A.Y., 28n8 Shell, 172 Shen, T., 116n32 Shenhua, 222n8 Shenzhen, 104, 106–107 network charges, 118 Shenzhen Energy Sales and Services Company (SESS), 122, 124 Short-term operating reserve (STOR), 198 Shu, C., 124n55 Sichuan, 219 characteristics of, 220 electricity balances in, 221 electricity production mix in, 221 progress with power market implementation, 222 Siemens, 15 Sinohydro, 12 Sinopec, 234 Sioshansi, F., 235n16 Sioshansi, R., 164n18 Sioshansi, S., 36n30 Slaughter and May, 65n81 SO, see System operators SOEs, see State-owned enterprises Solar PV, 69, 71

256 Index

South America, 2 competitive procurement of wholesale power, for regulated final customer groups, 59 independent system operators, 198 wholesale energy market, generation and demand in, 227 wholesale prices, 162 South China Energy Supervision Bureau, 131 SP Energy Networks, 159 SPC, see State Planning Commission; State Power Corporation Spot markets, 1, 119, 130, 137, 138, 161–163, 168, 169, 207 Chinese context, 46–47 general reform experience, 46 theoretical significance, 44–45 SSE, see Scottish and Southern Energy (SSE) Power Distribution Staffell, I., 206n70 Stakeholders of electricity supply industry generators, 237 grid companies, 238 policy-makers, 236 regulators, 237 retailers, 237 State Administration for Market Regulation (SAMR), 66 State Asset Holding Company, 65 State control, 40, 44 State Council, 140 State Electricity Regulatory Commission (SERC), 63

State Grid Corporation of China (SGCC), 10–12, 14, 30–32, 37, 42, 57, 67, 67n85, 87, 172, 208, 215, 222, 222n9, 227, 231, 232, 234 State of the Market Report for PJM, 225 State Power Corporation (SPC), 10 State Power Investment Corporation, 10, 222n8 State-owned Assets Supervision and Administration Commission (SASAC), 10, 12 State-owned enterprises (SOEs), 41, 73 Stern, J., 175 Stoft, S., 104, 114, 114n25, 115n31 STOR, see Short-term operating reserve STOR Runway, 198 Subsidy, 60, 69, 70, 73, 130, 161, 207 Sulphur/sulfur dioxide, 24, 76, 77, 79 Sun Art, 234 Sung, Y., 117n33 Supplemental Balancing Reserve (SBR), 200 Supportive secondary market arrangements efficient allocation, 49–54 regulated third-party access, 49–54 spot markets and ancillary services, 44–47 transmission capacity, 49–54 wholesale electricity markets, demand-side participation in, 47–49

 Index 

SWEB, 201 Sweeney, J.L., 29n14, 29n15, 162n14 System balancing charges, 190–200 System operation, 35, 55, 153, 173, 190, 196–198 System Operator Incentive Schemes, 195 System operators (SO), 114, 115, 127, 128, 157, 161–164, 166, 168, 169, 186, 190, 192–194, 195n54, 217 Chinese context, 36–40 general reform experience, 35–36 theoretical significance, 34–35 T

Tan, H., 81n117 Tariffs, 31, 33, 52, 58, 59, 73, 81, 89, 155, 170, 171, 173, 181, 186, 188, 203 Tax, 77, 86, 121, 130, 153, 154, 161, 164, 169, 182, 204–208, 217, 219 Taylor, N.W., 48n51 Taylor, S., 121n44, 167n30 TEC, see Transmission entry capacity TenneT, 187, 188 TGCs, see Tradable green certificates ‘There is no alternative’ (TINA) argument, 233 Three Gorges Dam, 235 13th Five Year Plan (FYP13), 8 Tianjin system operators, 36 TINA, see ‘There is no alternative’ argument

257

TNUoS, see Transmission use of system charges Tradable green certificates (TGCs), 70 Transition mechanisms Chinese context, 81–85 general reform experience, 80–81 theoretical significance, 80 Transmission, 25, 27–33, 38, 39, 41, 42, 49–58, 62, 63, 67, 68, 74–76, 104, 115, 117, 118, 130, 138, 154, 155, 157, 159, 160, 166, 168–170, 173, 175, 181, 182, 190, 191, 193, 196–198, 201, 203, 207 Transmission capacity, theoretical significance of, 49–52 Transmission charges, 3, 52n57, 55, 76, 117, 154, 155, 182–190 Transmission entry capacity (TEC), 185, 186 Transmission Network Reliability Incentive, 187 Transmission system operator (TSO), 164, 187, 188, 199 Transmission use of system charges (TNUoS), 184, 186, 202 demand charges, 185 generation charges, 185 TSO, see Transmission system operator 12th Five-Year Plan (2011–15), 64 Twomey, P., 69n87 TXU Europe, 121

258 Index U

UES, 41 UK, see United Kingdom UK Power Networks, 238 UKPN, 157, 159 Power Potential Project, 203n62 Ultra-high voltage (UHV), see High Voltage Direct Current (HVDC) United Kingdom (UK), 2, 3, 25, 26, 29n16, 138, 217 1987 General Election, 140 Competition and Markets Authority, 128 competitive procurement process, for low carbon generation, 71 decarbonisation and power sector, combining, 234 exposure to wholesale market prices, 127 Flexible Plug and Play project, 75 full implications of electric power, 231 industrial electricity prices, determination of, 153–208 market restructuring and ownership changes, 32 mitigation of social effects, 229, 230 Network Rail, 42 new energy market players, 120, 121 power sector reform without private ownership, 233 privatisation and monopolies, 41, 42 regulated network and competitive segment charges, 56, 58

regulatory agencies, for monopoly network charges regulation and monitoring competitive segments, 62, 63, 66n83, 67, 68 regulatory reporting, improvement of, 225, 226 renewables, cost reflective access terms for, 75 Royal Mail, 42 spot markets and ancillary services, 45n47, 46n48 State of the Energy Market report, 225 system operators, 34, 35 transition mechanisms, 81 transmission capacity, 52n57, 54 utility privatisation programme, 85 vertical separation and horizontal restructuring, 29 wholesale electricity markets, demand-side participation in, 48, 49 zonal transmission charges, 76 United States (US), 2, 25, 85, 216, 218 Californian electricity crisis of 2001, 231 competitive procurement of wholesale power, for regulated final customer groups, 59, 60n74 competitive procurement process, for low carbon generation, 71 electricity price, 85, 86 full implications of electric power, 231

 Index 

independent system operators, 198 industrial electricity price, 219 interprovincial market, creation of, 228, 229 learning from pilot markets, promoting, 227 policy-makers in, 236 power markets, 113 power sector reform in, 16–19 power sector reform without private ownership, 233 progressing power sector reform in, 232 regulated network and competitive segment charges, 55 regulatory reporting, improvement of, 225 State of the Markets Report, 225 system operators, 35, 36 transmission capacity, 53 vertical separation and horizontal restructuring, 28 wholesale electricity markets, demand-side participation in, 48, 49 wholesale energy market, generation and demand in, 227 US, see United States Utilisation, 38n37, 130, 224 V

Vertical separation, 26–29 general reform experience, 28–29 theoretical significance, 27–28 Viscusi, W.K., 77n105 Vona, F., 156n4

259

W

WACC, see Weighted average cost of capital Wales industrial electricity price, determination of, 157 system operators, 34, 35 Wang, C., 128n64 Wang, J., 49n53 Wang, Q., 15n18 Wang, W., 108n15 Wang, X., 37n33 Wei, Y., 38n36, 78n109 Weighted average cost of capital (WACC), 176 Wen, H., 103n4 Wenchuan earthquake (2008), 15 White, M.W., 35n26, 115, 115n30, 128n61, 169n35 Wholesale electricity markets, demand-side participation in Chinese context, 48–49 general reform experience, 48 theoretical significance, 47–48 Wholesale energy market, generation and demand in, 227 Wholesale prices, 161–169 Wilson, I.A.G., 206n70 Wilson, S., 140n93 Wind, 69, 71, 73, 74, 86, 88, 167, 189, 205 World Bank, 61, 61n76 WPD, 157, 159 X

Xu, Y.C., 7n5, 12n11, 15n16, 36n32, 60n75, 64n78, 85

260 Index Y

Yang, W., 109n20 Yang, Y., 140n93 Yao, M., 47n50 Yeo, Y., 64n79 Yu, H., 4n2 Yu, J.Q., 4n2, 132, 132n77 Yu, S., 78n111 Yuan, 79n112 Yudean, 132 Yunnan, 129, 137, 138 network charges, 182 system operators, 36 transmission capacity, 53 wholesale electricity markets, demand-side participation in, 49 Z

Zeng, L., 114, 114n27 Zhang, D., 140n93

Zhang, C., 41n42 Zhang, H., 108n17, 133n80 Zhang, O., 36n31, 76n102 Zhang, S.L., 49n53, 76n102, 129n66, 138n86 Zhe, Yao, 108n14 Zhejiang, 10, 219, 221–223 characteristics of, 220 electricity balances in, 221 electricity production ix in, 221 learning from pilot markets, promoting, 227 progress with power market implementation, 222 system operators, 39 Zheng, D., 47n50 Zheng, W., 53n62, 57n66 Zheng, X., 78n111, 81n118 Zhongfu, T., 37n34 Zhou, H., 53n60 Zhou, W., 47n50