World Energy Outlook 2013 - International Energy Agency

Loading...
Please note that this PDF is subject to specific restrictions that limit its use and distribution. The terms and conditions are available online at http://www.iea.org/ termsandconditionsuseandcopyright/

WORLD

ENERGY

OUTLOOK

2 0 1 3

WORLD ENERGY OUTLOOK 2013 In a world where big differences in regional energy prices impact competitiveness, who are the potential winners and losers? Huge volumes of oil are needed to meet growing demand and offset declines in existing fields. Where will it all come from? What could trigger a rapid convergence in natural gas prices between Asia, Europe and North America, and how would it affect energy markets? Is the growth in renewable energy self-sustaining and is it sufficient to put us on track to meet global climate goals? How much progress is being made in phasing-out fossil-fuel subsidies and expanding access to modern energy services to the world’s poor? The answers to these and many other questions are found in WEO-2013, which covers the prospects for all energy sources, regions and sectors to 2035. Oil is analysed in-depth: resources, production, demand, refining and international trade. Energy efficiency – a major factor in the global energy balance – is treated in much the same way as conventional fuels: Its prospects and contribution are presented in a dedicated chapter. And the report examines the outlook for Brazil’s energy sector in detail and the implications for the global energy landscape.

€150 (61 2013 13 1P1) ISBN: 978-92-64-20130-9

WORLD

ENERGY

OUTLOOK

2 0 1 3

INTERNATIONAL ENERGY AGENCY The International Energy Agency (IEA), an autonomous agency, was established in November 1974. Its primary mandate was – and is – two-fold: to promote energy security amongst its member countries through collective response to physical disruptions in oil supply, and provide authoritative research and analysis on ways to ensure reliable, affordable and clean energy for its 28 member countries and beyond. The IEA carries out a comprehensive programme of energy co-operation among its member countries, each of which is obliged to hold oil stocks equivalent to 90 days of its net imports. The Agency’s aims include the following objectives: „ Secure member countries’ access to reliable and ample supplies of all forms of energy; in particular, through maintaining effective emergency response capabilities in case of oil supply disruptions. „ Promote sustainable energy policies that spur economic growth and environmental protection in a global context – particularly in terms of reducing greenhouse-gas emissions that contribute to climate change. „ Improve transparency of international markets through collection and analysis of energy data. „ Support global collaboration on energy technology to secure future energy supplies and mitigate their environmental impact, including through improved energy efficiency and development and deployment of low-carbon technologies. „ Find solutions to global energy challenges through engagement and dialogue with non-member countries, industry, international organisations and other stakeholders.

IEA member countries: Australia Austria Belgium Canada Czech Republic Denmark Finland France Germany Greece Hungary Ireland Italy Japan Korea (Republic of) Luxembourg Netherlands New Zealand Norway Poland Portugal © OECD/IEA, 2013 Slovak Republic International Energy Agency Spain 9 rue de la Fédération Sweden 75739 Paris Cedex 15, France Switzerland www.iea.org Turkey United Kingdom Please note that this publication United States is subject to specific restrictions that limit its use and distribution. The terms and conditions are available online at http://www.iea.org/termsandconditionsuseandcopyright/

The European Commission also participates in the work of the IEA.

Foreword

This year’s World Energy Outlook, WEO-2013, should make us all stop and think. r. &aƟh irol and his dediĐated team at the /, supported ďy many others in the ŐenĐy and oǀer ϮϬϬ hiŐhlyͲƋualiĮed edžternal reǀieǁers, haǀe Őiǀen us another arresƟnŐ ǀieǁ oĨ the main deǀelopments in the enerŐy ǁorld oǀer the last year and draǁn out the impliĐaƟons Ĩor the Ĩuture. te haǀe ďeen priǀileŐed to ǁork this year ǁith many edžperts on ranjilian enerŐy to analyse that Đountry’s prospeĐts. te haǀe done the same ǁith Őoǀernments and edžternal edžperts ǁith respeĐt to ^outheast sia ʹ and haǀe puďlished our ĮndinŐs separately to respond to the Ɵmetaďle oĨ important reŐional disĐussions. &rom a ǁealth oĨ inĨormaƟon and analysis, / seleĐt some oĨ the key ĮndinŐs ǁhiĐh haǀe seinjed my aƩenƟon and ǁill Őuide my thinkinŐ durinŐ the ĐominŐ months͗ „ Kn the ďasis oĨ the intenƟons already edžpressed ďy Őoǀernments, enerŐy eĸĐienĐy is

set to ͞supply͟ more addiƟonal enerŐy than oil throuŐh to ϮϬϯϱ. nerŐy eĸĐienĐy is the only ͞Ĩuel͟ that simultaneously meets eĐonomiĐ, enerŐy seĐurity and enǀironmental oďũeĐƟǀes. „ ,alĨ the inĐrease in the ǁorld’s eleĐtriĐity output to ϮϬϯϱ Đomes Ĩrom reneǁaďles.

sariaďle sourĐes ʹ ǁind and solar ʹ make up a larŐe part oĨ the inĐrease. s inteŐraƟnŐ these ǀariaďle reneǁaďles Đan ďe Đompledž and Đostly, poliĐies to support their deployment need to ďe Đomplemented ďy aĐƟon on inĨrastruĐture deǀelopment and, in some Đases, market struĐture. „ Kn the ďaĐk oĨ liŐht ƟŐht oil output, the hnited ^tates is on the ǀerŐe oĨ ďeĐominŐ the

ǁorld’s larŐest oil produĐer and is ǁell on its ǁay to realisinŐ the meriĐan dream oĨ net enerŐy selĨͲsuĸĐienĐy. „ The Diddle ast, lonŐ thouŐht oĨ primarily as a supplier to ǁorld enerŐy markets, is

ďeĐominŐ a maũor enerŐy Đonsumer. 'roǁth in Diddle ast oil ĐonsumpƟon ďy ϮϬϯϱ enƟrely oīsets the reduĐƟon in ĐonsumpƟon in K urope. 'roǁth in Diddle ast Őas demand to ϮϬϯϱ, in aďsolute terms, is seĐond only to that oĨ hina. „ The ǁorld oĨ oil reĮninŐ is in transiƟon. Darkets are shiŌinŐ east, as demand Őroǁs

© OECD/IEA, 2013

in the deǀelopinŐ ǁorld and Ĩalls in K Đountries, ǁhile ĨeedstoĐk ĐhanŐes redeĮne the reƋuired ĐharaĐterisƟĐs oĨ reĮneries. nyone ǁith speĐial interest in one oĨ the Ĩuels ʹ espeĐially oil, ǁhiĐh ǁe analyse in depth this year ʹ ǁill Įnd their oǁn insiŐhts in the releǀant Đhapters. nerŐy eĸĐienĐy ʹ the ulƟmate alternaƟǀe Ĩuel ʹ is edžplored on an eƋual ďasis in its oǁn Đhapter. hapter ϴ, on ĐompeƟƟǀeness, inĨorms the ŐroǁinŐ deďate aďout the impliĐaƟons Ĩor industrial ĐompeƟƟǀeness oĨ diīerenĐes in enerŐy priĐes aĐross the reŐions oĨ the ǁorld.

Foreword

3

Our purpose in the WEO is oďũeĐƟǀe͗ to draǁ aƩenƟon to the Đourse on ǁhiĐh the enerŐy ǁorld is set and to point out the issues that arise. ut ǁe haǀe ideas Ĩor soluƟons and stand ready to disĐuss them ǁith all memďers oĨ the Đommunity touĐhed ďy the Ĩortunes and operaƟons oĨ the enerŐy seĐtor. This puďliĐaƟon is produĐed under my authority as džeĐuƟǀe ireĐtor oĨ the /.

© OECD/IEA, 2013

Maria van der Hoeven džeĐƵƟve ireĐƚor /nƚernaƟonaů nerŐLJ ŐenĐLJ

4

World Energy Outlook 2013

Acknowledgements

This study ǁas prepared ďy the ireĐtorate oĨ 'loďal nerŐy ĐonomiĐs oĨ the /nternaƟonal nerŐy ŐenĐy in ĐoͲoperaƟon ǁith other direĐtorates and oĸĐes oĨ the ŐenĐy. /t ǁas designed and directed by &aƟŚථBirol, hieĨ conomist oĨ the /. LauraථCozzi co-ordinated the analysis oĨ climate change, energy eĸciency and modelling͖ TimථGould co-ordinated the analysis oĨ oil, natural gas and ranjil͖ AmosථBromŚead co-ordinated analysis of ASEAN and fossil-fuel subsidies͖ CŚrisƟanථBesson co-ordinated the analysis of oil͖ anථorner co-ordinated the analysis of ranjil and energy access͖ MarĐo Baroni co-ordinated the poǁer and reneǁables analysis͖ WaǁeųථKleũarniŬ co-ordinated the analysis of coal and compeƟƟǀeness͖ TimurථGƺl co-ordinated the transport analysis. Other colleagues in the irectorate of 'lobal Energy Economics contributed to mulƟple aspects of the analysis and ǁere instrumental in deliǀering the study͗ AliථAlͲ^aīar ;ranjil, oilͿ͖ AlessandroථBlasi ;ASEAN, ranjilͿ͖ /anථCronsŚaǁ ;coal, natural gasͿ͖ CaƉellaථ&esƚa ;ranjil, oilͿ͖ MaƩŚeǁථ&ranŬ ;oil, poǁerͿ͖ ^ŚiŐeƚosŚiථ/ŬeLJama ;ASEAN, policiesͿ͖ Barƚoszථ:urŐa ;unconǀenƟonal gasͿ͖ &aďianථKħsiĐŬi ;energy eĸciency, petrochemicalsͿ͖ Soo-IlථKim ;industry, policiesͿ͖ CaƚurථKurniadi ;ASEAN͖ poǁerͿ͖ AƚsuŚiƚoථKurozumi ;assumpƟons, policiesͿ͖ :unŐථWooථLee ;fossil-fuel subsidies, buildingsͿ͖ Berƚrand MaŐnĠ ;compeƟƟǀeness, energyථeĸciencyͿ, CŚiaraථMarriĐĐŚi ;poǁer, reneǁablesͿ͖ KrisƟne WeƚrosLJan ;oil reĮning and tradeͿ͖ KaƚrinථSĐŚaďer ;buildings, reneǁablesͿ͖ NoraථSelmeƚ (fossil-fuel subsidies, energy accessͿ͖ SŚiŐeruථSueŚiro (industry, assumpƟonsͿ͖ TimurථToƉalŐoeŬĐeli (oil, naturalථgasͿ͖ :oŚannesථTrƺďLJ (poǁer, coalͿ͖ KeesථsanථNoorƚ (oil, naturalථgasͿ͖ BrenƚථWanner (Brazil, poǁerͿ͖ DavidථWilŬinson (poǁer, reneǁablesͿ͖ SŚuǁeiථŚanŐ (transportͿ. Sandra MooneLJ, MaŐdalena SanoĐŬa and MarLJZose Cleere proǀided essenƟal support. Dore details about the team can be found at www.worldenergyoutlook.org. Zoďerƚ Wriddle carried editorial responsibility.

© OECD/IEA, 2013

:ean hateau from the OE proǀided ǀaluable input to our macroeconomic modelling. The study also beneĮted from input proǀided by numerous /EA edžperts. /n parƟcular, Danuel Baritaud, Philippe Benoit, Ulrich Benterbusch, Toril Bosoni, Stéphanie Bouckaert, Adam Broǁn, Dichael ohen, &rancois uenot, arlos &ernandez Alǀarez, Araceli &ernandez Pales, Zebecca 'aghen, :ean-&rancois 'agné, :ean-zǀes 'arnier, Paolo &rankl, Antoine ,alī, tolf ,eidug, hrisƟna ,ood, idier ,oussin, :oerg ,usar, Aledž <Ƃrner, iane Dunro, DaƩheǁ Parry, aszlo sarro. Thanks go to the /EA’s ommunicaƟon and /nformaƟon Oĸce for their help in producing the Įnal report, and to Bertrand Sadin and Anne Dayne for graphics. ebra :ustus ǁas copy-editor. The special focus on Brazil beneĮted greatly from close co-operaƟon ǁith the goǀernment of Brazil. te are parƟcularly grateful to DĄrcio immermann, EdžecuƟǀe Secretary (eputy DinisterͿ at the Dinistry of Dines and Energy in Brazil, for his support throughout the study and for proǀiding the keynote address at our ǁorkshop in Zio de :aneiro, and to

Acknowledgements

5

&rancisco toũcicki, eputy EdžecuƟǀe Secretary, and many other senior oĸcials and edžperts at the Dinistry for their Ɵme and assistance. Sincere thanks also to Albert de Delo, irector 'eneral of the Electric Poǁer Zesearch ompany (Eletrobras EPE>Ϳ and to Dauricio TolmasƋuim, President of the Energy Zesearch ompany (EPEͿ for their ǀaluable insights and edžperƟse. te are grateful to Antonio astro, EdžecuƟǀe Danager for orporate Strategy at Petrobras, for his support and his special address to our workshop, and to the numerous senior edžperts from Petrobras who shared their ǀiews with the /EA team, including those that were seconded to the /EA oǀer the course of the analysis. Edmar Almeida (&ederal Uniǀersity of Zio de :aneiroͿ, Nick Butler (ondonͿ, Treǀor Dorgan (Denecon onsulƟngͿ and UŒur Pcal (Ocal Energy onsulƟngͿ proǀided ǀaluable input into the analysis. The work could not haǀe been achieǀed without the substanƟal support and co-operaƟon proǀided by many goǀernment bodies, organisaƟons and energy companies worldwide, notably͗ epartment of Zesources, Energy and Tourism, Australia͖ Dinistry of limate, Energy and Building, enmark͖ Enel͖ Eni͖ The /nsƟtute of Energy Economics, :apan͖ The :apan 'as AssociaƟon͖ Dinistry of Economy, Trade and /ndustry, :apan͖
Workshops A number of speciĮc workshops and meeƟngs were organised to get essenƟal input to this study. The workshop parƟcipants haǀe oīered ǀaluable new insights, feedback and data for this analysis. „ Zedrawing the Energy-limate Dap, Paris͗ ϴ Darch ϮϬϭϯ „ UnconǀenƟonal 'as &orum, Paris͗ ϮϮ Darch ϮϬϭϯ „ Brazil Energy Outlook, Zio de :aneiro͗ ϭϭ April ϮϬϭϯ „ The &uture of the Tight >iƋuids ZeǀoluƟon, Paris͗ ϯϬ April ϮϬϭϯ „ Southeast Asia Energy Outlook, Bangkok͗ ϳ Day ϮϬϭϯ

© OECD/IEA, 2013

„ /nternaƟonal Energy torkshop, Paris͗ ϭϵ-Ϯϭ :une ϮϬϭϯ

6

World Energy Outlook 2013

IEA Energy Business Council Special thanks go to the companies that parƟcipated in the meeƟng of the /EA Energy Business ouncil (EBͿ and ũoint session with the /EA 'oǀerning Board in :une 2013, which proǀided signiĮcant inputs, comments and insights to this study. The EB brings together many of the world’s largest companies inǀolǀed in all aspects of the energy industry, ranging from uƟliƟes, oil, natural gas and coal companies to automobile and appliance manufacturers, wind and solar producers and Įnancial insƟtuƟons. &urther details may be found at www.iea.org/energybusinesscouncil.

Peer reviewers

© OECD/IEA, 2013

Dany internaƟonal edžperts proǀided input, commented on the underlying analyƟcal work and reǀiewed early draŌs of the chapters. Their comments and suggesƟons were of great ǀalue. They include͗ Emmanuel Ackom Prasoon Agarwal Ali Aissaoui Dorten Albaek Edmar Almeida Ricardo Donte Alto Harald Andruleit Darco Annunziata Oliǀier Appert Andre Arauũo Gregſrio AraƷũo Jorge Arbache Laura Atkins Peter Aukamp Christopher Baker Amit Bando Paul Baruya Georg Bäuml Carmen Becerril Chris Beddoes
Acknowledgements

UNEP Risø Centre 'lobal 'reen 'rowth /nsƟtute AP/CORP Vestas &ederal Uniǀersity of Rio de :aneiro (U&R:Ϳ tood Dackenzie BGR General Electric /&P Shell Petrobras Brazilian NaƟonal eǀelopment Bank (BNESͿ Hart Energy Deutsche Bank Department of Climate Change and Energy Eĸciency, Australia /nternaƟonal Partnership for Energy Eĸciency CooperaƟon IEA Clean Coal Centre Volkswagen Acciona Group Europia Schlumberger AREVA Asian Deǀelopment Bank Energy Academy Europe torld Health OrganizaƟon European Council for an Energy EĸcientථEconomy (ECEEEͿ ED& Center for Energy, Darine TransportaƟon and Public Policy

7

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18

© OECD/IEA, 2013

Nigel Bruce Dichael Buĸer Nick Butler Ron Cameron Guy Caruso Arnaldo de Carǀalho Antonio Castro Surya Prakash Chandak Carla Cohen Alan Copeland João Coral John Corben Eduardo Correia Jostein Dahl
8

torld Health OrganizaƟon Glencore
World Energy Outlook 2013

© OECD/IEA, 2013

José Goldemberg Rainer GƂrgen Hugo Gosmann Andrew Gould Andrii Gritseǀskyi Howard Gruenspecht Cecilia Gunnarsson Edgard Habib Darshall Hall tenke Han Dargareta Drzeniek Hanouz Ali Hasanbeigi Douglas Heath James Henderson James HewleƩ Dasazumi Hirono Gilberto Hollauer Tom Howes Noé Van Hulst Dichio Ikeda Viũay Iyer Anil Jain Iǀan Jaques James Jensen Jan-Hein Jesse Balazs Jozsa Darianne
Acknowledgements

Uniǀersity of São Paulo &ederal Dinistry of Economy and Technology, Germany Dinistry of Dines and Energy, Brazil BG Group InternaƟonal Atomic Energy Agency Department of Energy, United States Volǀo Cheǀron BG Group Energy Research InsƟtute of NDRC, China torld Economic &orum Lawrence Berkeley NaƟonal Laboratory Natural Resources Canada Odžford InsƟtute of Energy Studies Department of Energy, United States The Japan Gas AssociaƟon Dinistry of Dines and Energy, Brazil European Commission Energy Academy Europe Jy Nippon Oil Θ Energy CorporaƟon The torld Bank Planning Commission, India The torld Bank Jensen Associates Clingendael InsƟtute European Commission ConocoPhillips Ditsubishi CorporaƟon Dinistry of Economy, Trade and Industryථ(DETIͿ, Japan Energy Intelligence Group InternaƟonal Aluminium InsƟtute Danish Energy Agency Europe NL Agency InsƟtute of Energy Economics, Japan Center for Strategic and InternaƟonal Studies Yueiroz Galǀão EdžploraƟon and ProducƟon Iberdrola Planning Commission, Pakistan Energy Intelligence Group Global Buildings Performance Network Brazilian AssociaƟon of Electricity Distributors (ABRADEͿ Eskom

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 9

© OECD/IEA, 2013

Dark Lewis tenge Liu Xiaoli Liu Philip Lowe João de Luca Breno Lyro Joe Dastrangelo Ritu Dathur Dichael DcNeil Dichael Dellish Albert de Delo Pedro Derino Lawrence Detzroth Gergana Diladinoǀa Russel Dills Ryo Dinami TaƟana Ditroǀa Duzi Dkhize
Deutsche Bank China Coal InformaƟon InsƟtute Energy Research InsƟtute of NDRC, China European Commission Brazilian InsƟtute of Oil and Gas and Biofuels Petrobras General Electric The Energy and Resources InsƟtute (TERIͿ Lawrence Berkeley NaƟonal Laboratory Department of Energy, United States Dinistry of Dines and Energy, Brazil Repsol Arch Coal European Commission Dow Chemical Dinistry of Economy, Trade and Industryථ(DETIͿ, Japan Energy Research InsƟtute of the Russian Academy of Sciences Department of Energy, South Africa Uniǀersity of Staǀanger Business School United NaƟons Industrial Deǀelopment OrganizaƟon Enel CiƟgroup American Council for an Energy-Eĸcient Economy (ACEEEͿ J-Power Ditsubishi CorporaƟon Noǀozymes &ederal Uniǀersity of ItaũubĄ (UNI&EIͿ, Brazil Rystad Energy Natural Resources Canada Dichelin Dinistry of Dines and Energy, Brazil EdždžonDobil Ditsubishi CorporaƟon InternaƟonal InsƟtute for Applied Systems Analysis (IIASAͿ PSR, Brazil Dinistry of Dines and Energy, Brazil Siemens Tata Steel Edison Independent consultant European Commission Department of Energy and Climate Change, United
Brian RickeƩs Jeīrey Rinker Rahim Saad
© OECD/IEA, 2013

Danfred Schuckert Robert Schwiers Radia Sedaoui Hana-Duriel SeƩeboun Adnan Shihab-Eldin Daria Sicilia Salǀadores Pierre Sigonney Nelson Silǀa Baoguo Shan Stephan Singer Christopher Snary João Souto Eduardo de Sousa Elias de Souza Robert Spicer Leena Sriǀastaǀa John Staub Ron Steenblik Osǀaldo Stella Jonathan Stern Samya Beidas-Strom Dichael Stoppard Ulrik Stridbaek Greg Stringham Cartan Sumner GƂran Sǀensson Renata Szczerbacki Philippine de T’Serclaes Dinoru Takada Nogami Takayuki

Acknowledgements

EURACOAL ODV BG Group European Commission Odžford InsƟtute for Energy Studies Global tind Energy Council &ederal Uniǀersity of Rio de Janeiro (U&RJͿ Qatar Petroleum RtE BGR Dinistry for the Enǀironment, Nature ConserǀaƟon and Nuclear Safety, Germany Daimler AG Cheǀron Gas EdžporƟng Countries &orum &< Group
11

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18


J-POtER &ormer EdžecuƟǀe Director, IEA E.ON Petrobras Shell Energy Research Company (EPE), Brazil SUE< Dinistry of &oreign Aīairs and Trade, New ealand Trina Solar Center for Strategic and InternaƟonal Studies Statoil Pledžus Energy Uniǀersity of Campinas, Brazil SABIC RtE IHS CERA Cardiī Business School Sanford C. Bernstein Limited BAS& Independent edžpert Peabody Energy Department of Resources, Energy and Tourism, Australia Consol Dinistry of Dines and Energy, Brazil Shell Dinistry of Enǀironment, Poland United NaƟons &oundaƟon Chinese Academy of Sciences CiƟ Rio Tinto ICIS ConsulƟng

© OECD/IEA, 2013

The indiǀiduals and organisaƟons that contributed to this study are not responsible for any opinions of ũudgments contained in this study. All errors and omissions are solely the responsibility of the IEA.

12

World Energy Outlook 2013

Commenƚs and ƋuesƟons are ǁelĐome and sŚould ďe addressed ƚo͗

1

Dr. &aƟh Birol Chief Economist Director, Directorate of Global Energy Economics InternaƟonal Energy Agency 9, rue de la &édéraƟon 75739 Paris Cededž 15 &rance Telephone͗ Email͗

2 3 4

(33-1) 4057 6670 weoΛiea.org

5

Dore informaƟon about the World Energy Outlook is aǀailable at www.worldenergyoutlook.org.

6 7 8 9 10 11 12 13 14 15 16

© OECD/IEA, 2013

17 18 Acknowledgements

13

T A B L E

PART A GLOBAL ENERGY TRENDS

© OECD/IEA, 2013

O F C O N T E N T S

PART B BRAZIL ENERGY OUTLOOK

PART C OUTLOOK FOR OIL MARKETS

ANNEXES

© OECD/IEA, 2013

SCOPE AND METHODOLOGY

1

GLOBAL ENERGY TRENDS TO 2035

2

NATURAL GAS MARKET OUTLOOK

3

COAL MARKET OUTLOOK

4

POWER SECTOR OUTLOOK

5

RENEWABLE ENERGY OUTLOOK

6

ENERGY EFFICIENCY OUTLOOK

7

ENERGY AND COMPETITIVENESS

8

THE BRAZILIAN ENERGY SECTOR TODAY

9

PROSPECTS FOR BRAZIL’S DOMESTIC ENERGY CONSUMPTION

10

BRAZILIAN RESOURCES AND SUPPLY POTENTIAL

11

IMPLICATIONS OF BRAZIL’S ENERGY DEVELOPMENT

12

FROM OIL RESOURCES TO RESERVES

13

PROSPECTS FOR OIL SUPPLY

14

PROSPECTS FOR OIL DEMAND

15

IMPLICATIONS FOR OIL REFINING AND TRADE

16

ANNEXES

Foreword Acknowledgements Executive Summary

3 5 23

Part A: GLOBAL ENERGY TRENDS

31

1

2

© OECD/IEA, 2013

3

16

Scope and methodology

33

Scope of report Introducing the scenarios Building blocks of the scenarios Economic growth Population and demographics Energy prices Oil prices Natural gas prices Steam coal prices Carbon markets Technology

34 34 37 37 40 43 45 45 47 49 51

Global energy trends to 2035

55

Overview of energy trends by scenario Energy trends in the New Policies Scenario Energy demand Energy supply Inter-regional energy trade Implications for energy-related CO2 emissions Topics in focus Energy trends in Southeast Asia Modern energy for all Energy subsidies

56 59 59 71 76 79 83 84 87 93

Natural gas market outlook

99

Global overview Demand Regional trends Sectoral trends Production Resources and reserves Production trends Focus on unconventional gas Trade, pricing and investment Inter-regional trade Pricing of internationally traded gas Investment

100 102 102 105 107 107 108 115 123 123 128 136 World Energy Outlook 2013

4

5

© OECD/IEA, 2013

6

Coal market outlook

139

Overview Demand Regional trends Sectoral trends Supply Resources and reserves Production Trade Costs and investment Pricing of internationally traded coal Regional insights China United States India Australia ASEAN

140 143 143 145 147 147 148 150 152 154 156 156 159 163 165 166

Power sector outlook

169

Introduction Electricity demand Electricity supply Capacity retirements and additions Fossil-fuelled generation Nuclear power Renewables Transmission and distribution CO2 emissions Investment Electricity prices Residential Industry

170 171 174 176 181 186 188 189 190 191 192 194 195

Renewable energy outlook

197

Recent developments Renewables outlook by scenario Renewables outlook by use in the New Policies Scenario Power generation Biofuels Heat Focus on power generation from variable renewables Wind power Solar photovoltaics

198 199 201 201 204 207 208 209 210

Table of Contents

17

7

© OECD/IEA, 2013

8

18

Implications for electricity systems and markets Competitiveness and unit costs Bioenergy Investment Subsidies

212 217 221 224 225

Energy efficiency outlook

231

Introduction Current status of energy efficiency Recent progress Recent policy developments Recent sectoral trends The outlook for energy efficiency Trends by region Trends by sector Investment in energy efficiency Broader benefits Energy imports Impact on total household expenditure Local air pollution CO2 emissions

232 234 234 237 240 241 242 245 254 256 256 257 259 259

Energy and competitiveness

261

Energy and international competitiveness Energy price disparities Just how big are regional disparities in energy prices? How are regional energy price disparities set to evolve? Energy and industrial competitiveness Why do energy price disparities affect industrial competitiveness? How does carbon pricing affect industrial competitiveness? Focus on chemicals Focus on iron and steel Focus on refining The outlook for industrial energy and competitiveness Focus on the outlook for chemicals Energy and economic competitiveness What is the impact of energy price disparities on overall trade balances? What is the impact of energy price disparities on household income? Energy competitiveness and policy implications

262 267 267 271 275 275 282 283 285 287 287 291 292 294 296 297

World Energy Outlook 2013

Part B: BRAZIL ENERGY OUTLOOK 9

10

© OECD/IEA, 2013

11

301

The Brazilian energy sector today

303

Introducing Brazil’s energy sector Domestic energy trends Power sector Bioenergy Oil and gas Energy-related CO2 emissions and energy efficiency Regional and global interactions Projecting future developments The building blocks

304 306 309 312 315 318 319 321 322

Prospects for Brazil’s domestic energy consumption

329

Domestic energy consumption trends Outlook for the power sector Electricity demand Electricity generation Transmission and distribution Investment Outlook for other energy-consuming sectors Industry Transport Buildings Other sectors (agriculture, non-energy use) Outlook by fuel Oil products Natural gas Renewables Coal

330 332 332 335 340 341 344 344 348 351 354 354 354 355 358 359

Brazilian resources and supply potential

361

Oil and gas Resources and reserves Oil production Gas production Renewables Hydropower Biofuels Other bioenergy Wind Solar

362 362 367 378 383 383 387 391 391 393

Table of Contents

19

12

Other fuels Coal Nuclear

394 394 395

Implications of Brazil’s energy development

397

Context for Brazilian energy development Energy and the Brazilian economy Pricing Revenues Energy trade and security Regional co-operation Brazil and international oil and gas markets Brazil and international biofuels trade Energy and the environment Energy-related emissions Energy efficiency

398 398 403 403 405 405 407 410 411 413 415

Part C: OUTLOOK FOR OIL MARKETS 13

© OECD/IEA, 2013

14

20

419

From oil resources to reserves

421

Classifying oil Conventional oil Resources Reserves Enabling technologies: focus on enhanced oil recovery Unconventional oil Resources Reserves Enabling technologies Supply costs

422 425 425 430 437 446 446 451 452 453

Prospects for oil supply

457

Global oil supply trends Decline rate analysis The importance of decline Decline rates for conventional oil Decline rates for unconventional oil Implications for future production

458 459 459 462 466 468

World Energy Outlook 2013

15

16

Oil production by type Focus on light tight oil Oil production by region Non-OPEC OPEC Supply trends and potential implications for prices A Low Oil-Price Case Upstream industry structure Investment

471 474 479 479 483 488 490 493 495

Prospects for oil demand

501

Global oil demand trends Oil demand by region Focus on the Middle East Oil demand by sector Transport Industry Other sectors Oil demand by product

502 503 505 509 511 521 528 528

Implications for oil refining and trade

535

Making the connection between oil demand and supply Products bypassing the refining sector Natural gas liquids The refining sector Global refining outlook Refining sector outlook by region Oil trade

536 537 537 540 542 546 561

ANNEXES

© OECD/IEA, 2013

Annex A. Annex B. Annex C. Annex D.

567 Tables for Scenario Projections Policies and measures by scenario DeĮniƟons References

Table of Contents

569 645 657 671

21

List of figures Part A: GLOBAL ENERGY TRENDS Figures for Chapter 1: Scope and methodology

1.1 1.2 1.3 1.4

Primary energy demand and GDP GDP per capita by region Natural gas prices by region in the New Policies Scenario Current and proposed schemes that put a price on carbon

38 43 46 50

Figures for Chapter 2: Global energy trends to 2035

2.1 World primary energy demand and related CO2 emissions by scenario 57 2.2 Change in world primary energy demand by scenario, 2011-2035 59 2.3 Primary energy demand and energy intensity in the New Policies Scenario 60 2.4 Change in oil demand in selected regions in the New Policies Scenario 62 2.5 World primary energy demand by fuel in the New Policies Scenario 63 2.6 Share of world primary energy demand by region 65 2.7 Primary energy demand in selected regions and the share of global growth in the New Policies Scenario 67 2.8 Share of the growth in world primary energy demand by region in the New Policies Scenario 68 2.9 Change in energy demand by sector and fuel in the New Policies Scenario, 2011-2035 70 2.10 Fossil energy resources by type 72 2.11 Change in production by fuel in selected regions in the New Policies Scenario 74 2.12 Net oil and gas import/export shares in selected regions in the New Policies Scenario 77 2.13 Change in world energy-related CO2 emissions by policy measure in the 4-for-2 °C Scenario 81 2.14 Energy-related CO2 emissions by region in the New Policies Scenario 82 2.15 Energy-related CO2 emissions per capita and CO2 intensity in selected regions in the New Policies Scenario 83 2.16 Energy in Southeast Asia 85 2.17 Shares of population with access to electricity and clean cooking facilities by region in the New Policies Scenario 92 2.18 Economic value of fossil-fuel consumption subsidies by fuel for top 25 countries, 2012 95 2.19 Rates of fossil-fuel consumption subsidies in 2012 and recent developments in selected countries 97

© OECD/IEA, 2013

Figures for Chapter 3: Natural gas market outlook

3.1 3.2 3.3

Natural gas demand and production growth in selected regions, 2005-2012 Natural gas demand in selected regions in the New Policies Scenario World natural gas demand by sector in the New Policies Scenario

101 102 106

World Energy Outlook 2013

3.4 Change in annual natural gas production in selected countries in the New Policies Scenario 110 3.5 Remaining unconventional gas resources in selected regions, end-2012 116 3.6 Growth in unconventional gas production by type in selected regions in the New Policies Scenario 118 3.7 Unconventional gas production by selected country in the New Policies Scenario 120 3.8 European Union natural gas supply and demand balance in the New Policies Scenario 125 3.9 China natural gas supply and demand balance in the New Policies Scenario 126 3.10 Estimated shares of internationally traded gas by type of pricing mechanism 129 3.11 Regional gas prices in the New Policies Scenario and in the Gas Price Convergence Case 134 3.12 Differences in gas consumption between the Gas Price Convergence Case and the New Policies Scenario 135 3.13 Change in import volumes and imports bills for selected regions in the Gas Price Convergence Case, relative to the New Policies Scenario, 2035 136 3.14 Cumulative investment in natural gas supply infrastructure by region in the New Policies Scenario 137 Figures for Chapter 4: Coal market outlook

4.1 Incremental world coal demand, historical and by scenario 4.2 Coal demand by key region in the New Policies Scenario 4.3 Coal demand by key sector in the New Policies Scenario 4.4 Reserves and cumulative production by major coking and steam coal producers in the New Policies Scenario 4.5 FOB cash costs for seaborne steam coal exports, 2012 4.6 Quarterly indices for IEA crude oil and steam coal prices 4.7 China’s electricity generation in the New Policies Scenario 4.8 Coal demand in China and the rest of the world by major sector in the New Policies Scenario 4.9 Average costs of steam coal delivered to coastal China, 2012 4.10 US production cash costs for domestic steam coal, 2012 4.11 US net exports of coal in the New Policies Scenario 4.12 Major net importers of coal in the New Policies Scenario 4.13 Australian coal exports by type in the New Policies Scenario 4.14 ASEAN coal balance in the New Policies Scenario

140 143 147 148 153 154 157 157 159 160 162 164 165 167

© OECD/IEA, 2013

Figures for Chapter 5: Power sector outlook

5.1 World electricity demand by scenario relative to electricity demand assuming no change in electricity intensity 5.2 Electricity demand per capita in selected regions as a share of the OECD average in the New Policies Scenario 5.3 Electricity generation by source in the New Policies Scenario 5.4 Installed capacity by source in the New Policies Scenario Table of Contents

172 174 176 177

5.5 Age profile of installed thermal capacity by region, end-2012 5.6 Power generation gross capacity additions and retirements by selected region in the New Policies Scenario, 2013-2035 5.7 Power capacity changes in selected regions in the New Policies Scenario 5.8 Coal-fired power generation by region in the New Policies Scenario 5.9 Share of coal-fired power generation by technology and average efficiency in selected regions in the New Policies Scenario 5.10 Gas-fired power generation by selected region in the New Policies Scenario 5.11 Electricity generating costs for coal and gas by selected region and for 2008-2012 fuel prices 5.12 Nuclear power installed capacity by region in the New Policies Scenario 5.13 Renewables-based power generation and share of total generation by region in the New Policies Scenario 5.14 Existing and additional kilometres of transmission and distribution lines by selected region in the New Policies Scenario 5.15 CO2 emissions intensity in the power sector and electricity generation by region in the New Policies Scenario 5.16 Power sector cumulative investment by type and region in the New Policies Scenario, 2013-2035 5.17 Average residential electricity prices (excluding taxes) by region and cost component in the New Policies Scenario 5.18 Average industry electricity prices (excluding taxes) by region and cost component in the New Policies Scenario

177 180 181 182 183 184 185 187 188 189 191 192 194 195

© OECD/IEA, 2013

Figures for Chapter 6: Renewable energy outlook

6.1 Renewable energy share in total primary energy demand by category and region in the New Policies Scenario, 2011 and 2035 201 6.2 Incremental electricity generation from renewables in selected regions, 2011-2035 203 6.3 Average annual increases in renewables-based capacity by region in the New Policies Scenario 203 6.4 Cumulative global renewables-based capacity additions and retirements by technology in the New Policies Scenario, 2013-2035 204 6.5 Share of renewables in heat production in the residential sector for selected regions in the New Policy Scenario 207 6.6 Installed wind power capacity by region in the New Policies Scenario 210 6.7 Installed solar PV capacity by region in the New Policies Scenario 211 6.8 Indicative hourly electricity demand and residual electricity demand with expanding deployment of solar PV 214 6.9 Shares of wind and solar power capacity and generation in the New Policies scenario 215 6.10 Indicative breakeven costs of residential solar PV using the “grid parity” and “cost parity” approaches 219

World Energy Outlook 2013

6.11 Renewable electricity production costs relative to the wholesale prices for selected technologies and regions in the New Policies Scenario 6.12 World bioenergy use by sector in the New Policies Scenario 6.13 Biofuels demand and production in selected regions 6.14 Cumulative investment in renewables-based power generation capacity, 2013-2035 6.15 Global renewable energy subsidies by source in the New Policies Scenario 6.16 Global subsidies for renewable electricity generation and generation by source in the New Policies Scenario, 2013-2035 6.17 Renewables-based generation subsidies by source and selected region in the New Policies Scenario

© OECD/IEA, 2013

Figures for Chapter 7: Energy efficiency outlook

220 222 223 225 227 228 229

7.1 Proportion of long-term economic energy efficiency potential achieved in the New Policies Scenario, 2012-2035 232 7.2 Primary energy intensity levels and trends in selected regions, 2012 235 7.3 Annual relative change in global primary energy intensity by driver, 1990-2012 237 7.4 Energy intensity change by sector and region, 2005-2012 240 7.5 Change in global primary energy demand by category in the New Policies Scenario relative to the Current Policies Scenario 241 7.6a Energy intensity by region in the New Policies Scenario 243 7.6b Primary energy savings by region in the New Policies Scenario relative to the Current Policies Scenario in 2035 243 7.7 Primary energy savings from energy efficiency by fuel and sector in the New Policies Scenario relative to the Current Policies Scenario in 2035 245 7.8 Energy intensity reduction in the iron and steel sector by type of improvement, 2011-2035 247 7.9 Fuel savings from energy efficiency in road transport in the New Policies Scenario relative to the Current Policies Scenario 249 7.10a Energy consumption by end-use in households by region in the New Policies Scenario, 2011 and 2035 253 7.10b Final energy savings in households in the New Policies Scenario relative to the Current Policies Scenario by region, 2035 253 7.11 Average annual energy efficiency investment by scenario and sector 254 7.12 Payback periods for selected technologies and regions, 2013 255 7.13 Avoided import bills from energy efficiency in the New Policies Scenario relative to the Current Policies Scenario, 2035 256 7.14 Economic impacts of energy efficiency 257 7.15 Change in annual per-capita household spending on energy and non-energy goods and services in the New Policies Scenario relative to the Current Policies Scenario 259 7.16 World energy-related CO2 emissions abatement in the New Policies Scenario relative to the Current Policies Scenario 260

Table of Contents

Figures for Chapter 8: Energy and competitiveness

8.1 World energy use in industry by fuel in the New Policies Scenario 265 8.2 Ratio of Japanese and European natural gas import prices to United States natural gas spot price 267 8.3 Ratio of OECD coking to steam coal prices and Asian to European steam coal prices 268 8.4 Industrial energy prices including tax by fuel and region, 2012 269 8.5 Average industrial energy prices including tax by region 270 8.6 Ratio of Japanese and European natural gas import prices to United States natural gas spot prices in the New Policies Scenario 272 8.7 Industrial electricity prices by region and cost component in the New Policies Scenario 273 8.8 Ratio of European Union, Japanese and Chinese to US industrial electricity prices including tax in the New Policies Scenario 274 8.9 Industrial energy intensity by sub-sector and region, 2011 275 8.10 Share of energy in total production costs by sub-sector, 2011 276 8.11 Share of energy in total material costs in the United States 278 8.12 Value of the US dollar vis-à-vis other major currencies 281 8.13 Sensitivity of US industrial total material costs to CO2 prices, 2011 283 8.14 Share of energy in total material costs for selected chemical products in the United States, 2012 284 8.15 Historical and planned ethylene capacity additions by region 285 8.16 World incremental energy demand by industrial sub-sector and fuel in the New Policies Scenario 288 8.17 Regional shares of global export market value of energy-intensive industries in the New Policies Scenario 289 8.18a Regional shares of global export market and growth in export values by selected sector in the New Policies Scenario – Chemicals 291 8.18b Regional shares of global export market and growth in export values by selected sector in the New Policies Scenario – Non-ferrous metals 291 8.19 Compound average annual change in chemicals energy use and production by region in the New Policies Scenario 292 8.20 Overall trade balance including energy trade by region 294 8.21 Fossil fuel net import bills by region in the New Policies Scenario 295 8.22 Share of energy expenditures in household income by region in the New Policies Scenario 296 8.23 Compound average annual change in industrial production, efficiency and energy demand by scenario, 2011-2035 299

© OECD/IEA, 2013

Part B: BRAZIL ENERGY OUTLOOK Figures for Chapter 9: The Brazilian energy sector today

9.1 Share of renewables in total primary energy demand in selected regions, 2011 9.2 Energy map of Brazil

304 305

World Energy Outlook 2013

9.3 Brazil primary energy demand and GDP growth 9.4 Changes in income distribution in Brazil 9.5 Brazil domestic energy balance, 2011 9.6 Brazil final energy consumption by fuel in selected sectors 9.7 Brazil electricity supply by source 9.8 Brazil consumption of gasoline and ethanol in road transport 9.9 Evolution of Brazil’s proven oil and gas reserves 9.10 Brazil oil production and domestic demand 9.11 Brazil greenhouse-gas emissions by source 9.12 Energy intensity of GDP in Brazil as a share of selected regional and global averages 9.13 Brazil self-sufficiency in natural gas and selected oil products 9.14 Brazil’s energy policy and regulatory institutions Figures for Chapter 10: Prospects for Brazil’s domestic energy consumption

10.1 Brazil GDP and primary energy demand by scenario 10.2 Primary energy mix in Brazil and the world in the New Policies Scenario 10.3 Brazil electricity supply and demand by sector in the New Policies Scenario 10.4 Brazil power generation capacity additions in the New Policies Scenario 10.5 Brazil power generation by source in the New Policies Scenario 10.6 Brazil energy storage potential from hydropower reservoirs in compared with total generation in the New Policies Scenario 10.7 Brazil indicative monthly variations in power generation by source 10.8 Brazil’s electricity sector 10.9 Brazil average annual investment in the power sector in the New Policies Scenario 10.10 Brazil change in energy demand in selected energy-intensive manufacturing in the New Policies Scenario, 2011-2035 10.11 Average natural gas and electricity prices to industry by component 10.12 Brazil road-transport fuel demand by type in the New Policies Scenario 10.13 Share of domestic freight transport by mode in selected regions 10.14 Brazil change in energy demand in the buildings sector in Brazil in the New Policies Scenario, 2011-2035 10.15 Brazil natural gas demand by sector in the New Policies Scenario 10.16 Brazil consumption of non-hydro renewable energy by sector in the New Policies Scenario

© OECD/IEA, 2013

Figures for Chapter 11: Brazilian resources and supply potential

11.1 11.2 11.3 11.4 11.5 11.6 11.7

Main hydrocarbon basins in Brazil Global discoveries of super-giant oil fields Global deepwater oil output by company Brazil oil production by basin in the New Policies Scenario FPSO deployment in the New Policies Scenario and in the High Brazil Case Evolution of local content requirements in Brazil Main oil and gas fields and infrastructure in the Santos and Campos basins

Table of Contents

307 307 308 309 310 314 315 316 318 319 320 322 330 331 333 337 338 338 339 341 342 346 347 349 351 352 356 358 362 364 368 369 372 374 375

11.8 11.9 11.10 11.11 11.12 11.13 11.14 11.15 11.16

Brazil gas production in the New Policies Scenario Santos Basin gas production for different gas reinjection rates Brazil hydropower resources by river basin Brazil hydropower potential by classification of suitability Seasonal variation of selected renewable resources in Brazil Brazil agricultural land assessed as suitable for sugarcane production Biofuels production in selected regions in the New Policies Scenario Wind power capacity and capacity factors by country, 2020 and 2035 Top ten holders of uranium resources

379 381 384 385 386 388 388 392 396

Figures for Chapter 12: Implications of Brazil’s energy development

12.1 Average annual investment in Brazil’s energy supply infrastructure in the New Policies Scenario 399 12.2 Brazil share of installed deepwater subsea equipment and FPSOs in the New Policies Scenario, 2012 and 2020 400 12.3 Implications of different plateau production levels for the year in which 50% of Brazil’s oil resources are depleted 402 12.4 Oil export revenue as a share of GDP in selected countries in the New Policies Scenario 404 12.5 Major contributors to global oil supply growth in the New Policies Scenario, 2012-2035 407 12.6 Global deepwater oil production by region in the New Policies Scenario 408 12.7 Brazil oil balance in the New Policies Scenario 409 12.8 Brazil gas balance in the New Policies Scenario 410 12.9 CO2 per capita and CO2 intensity of GDP in selected regions in the New Policies Scenario 414 12.10 Energy-related CO2 emissions by sector, 2035 415 12.11 Brazil potential for energy efficiency savings by end-use sector relative to the New Policies Scenario, 2035 416

Part C: OUTLOOK FOR OIL MARKETS

© OECD/IEA, 2013

Figures for Chapter 13: From oil resources to reserves

13.1 Classification of oil resources 13.2 Classification of liquid fuels 13.3 Ultimately recoverable conventional crude oil resources and cumulative production required in the New Policies Scenario 13.4 Conventional crude oil resources by field size and year of discovery 13.5 Estimated conventional crude oil resources by field size 13.6 Observed discovery rates and average discovery size 13.7 Global exploration spending, 2000-2012 13.8 Evolution of published proven reserves for selected OPEC countries 13.9 Non-OPEC conventional crude oil proven reserves, 1980-2012

422 425 426 428 428 429 429 430 432

World Energy Outlook 2013

13.10 Ownership of 2P (“proven-plus-probable”) oil reserves by type of company, 2012 434 13.11 Distribution of proven-plus-probable reserves by region and type of company, 2012 434 13.12 Oil resources that are developed by scenario as a percentage of proven reserves 437 13.13 Estimated global EOR production by technology 442 13.14 Typical production profiles for LTO and EOR projects 443 13.15 EOR production by selected regions in the New Policies Scenario 445 13.16 Cumulative production versus remaining recoverable resources by type of unconventional oil in the New Policies Scenario 446 13.17 Supply costs of liquid fuels 454 13.18 World supply cost curves for 2013 and 2035 in the New Policies Scenario 455 13.19 Non-OPEC supply cost curves for 2013 and 2035 in the New Policies Scenario 456

© OECD/IEA, 2013

Figures for Chapter 14: Prospects for oil supply

14.1 Indicative illustration of decline phases and concepts 461 14.2 Observed year-on-year decline rate and weighted average CADR for conventional oil fields 463 14.3 Estimated difference between natural and observed decline rates in currently producing conventional fields 465 14.4 Typical production curve for a light tight oil well compared with a conventional oil well 467 14.5 Decline in production of conventional crude from currently producing fields in selected regions in the New Policies Scenario 469 14.6 Production that would be observed from all currently producing fields in the absence of further investment (excluding NGLs) 470 14.7 Projected evolution of natural decline rates in key regions in the New Policies Scenario, 2012-2035 471 14.8 Shares of world oil production by type in the New Policies Scenario 472 14.9 Production of NGLs in selected regions in the New Policies Scenario 473 14.10 Unconventional oil production in the New Policies Scenario 473 14.11 LTO production in selected countries in the New Policies Scenario 474 14.12 Projected LTO and NGLs production from unconventional plays in the United States in the New Policies Scenario 477 14.13 Change in oil production in selected countries in the New Policies Scenario, 2012-2035 486 14.14 Oil production changes by OPEC/non-OPEC grouping in the New Policies Scenario 489 14.15 Oil price and oil demand trajectories in the Low Oil-Price Case compared with the New Policies Scenario 491 14.16 Contributions to meeting the additional demand in the Low Oil-Price Case relative to the New Policies Scenario, 2035 492 14.17 Oil production by selected resource and company type, 2012 493 14.18 Oil production by company type in the New Policies Scenario 495

Table of Contents

14.19 Global share of oil production and investment by region in the New Policies Scenario 497 14.20 Worldwide upstream oil and gas investment and the IEA Upstream Investment Cost Index 499 Figures for Chapter 15: Prospects for oil demand

15.1 World oil demand and oil intensity by scenario 15.2 Growth in world oil demand by region in the New Policies Scenario, 2012-2035 15.3 Oil consumption subsidies and oil demand per capita by selected countries in the Middle East, 2012 15.4 Oil demand by sector in the Middle East 15.5 Electricity generating costs by technologies in the Middle East, 2015 15.6 Impact of fuel switching and efficiency on global oil demand in the New Policies Scenario 15.7 World oil demand for transport by sub-sector in the New Policies Scenario 15.8 PLDV vehicle fleet growth by region in the New Policies Scenario 15.9 Fuel mix in road-transport energy demand in the New Policies Scenario 15.10 Estimated payback periods of LNG-powered long-haul trucks in selected markets, 2011 and 2020 15.11 Natural gas demand for road transport by selected regions in the New Policies Scenario 15.12 Change of industrial oil demand (excluding feedstocks) by driver in the New Policies Scenario, 2012-2035 15.13 Production of high-value chemicals in the New Policies Scenario 15.14 Demand for oil as petrochemical feedstock by region in the New Policies Scenario 15.15 Simplified principal petrochemical product chains 15.16 Change in demand for oil by product in the New Policies Scenario, 2012-2035 15.17 LPG demand in the New Policies Scenario 15.18 Gasoline demand by region in the New Policies Scenario 15.19 Diesel demand by sector in the New Policies Scenario 15.20 World diesel demand by sector in the New Policies Scenario 15.21 Heavy fuel oil demand by sector in the New Policies Scenario

502 504 506 507 508 509 512 513 515 517 519 522 523 524 525 530 531 532 533 533 534

© OECD/IEA, 2013

Figures for Chapter 16: Implications for oil refining and trade

16.1 World oil production by quality in the New Policies Scenario 536 16.2a World liquids supply in the New Policies Scenario, 2012 and 2035 – 2012 538 16.2b World liquids supply in the New Policies Scenario, 2012 and 2035 – 2035 538 16.3 Routes to market for NGLs by process in the New Policies Scenario 539 16.4 NGLs product yields in the New Policies Scenario 540 16.5 Outputs from simple distillation versus final product demand 541 16.6 World refining capacity 542 16.7 Changes in refinery runs and changes in demand in the New Policies Scenario, 2012-2035 546

World Energy Outlook 2013

16.8 Selected regions shown by their net trade position in crude oil and oil products, 2012 547 16.9 European product demand split compared with the rest of the world, 2012 548 16.10 European dependence on trade for selected transport fuels in the New Policies Scenario 549 16.11 Share of imported crude in North American and European refinery runs 550 16.12 European refining capacity, demand and oil production in the New Policies Scenario 551 16.13 North American gasoline and diesel balances in the New Policies Scenario 553 16.14 North American trade of LPG and naphtha in the New Policies Scenario 554 16.15 Crude oil trade in selected Asian countries and regions in the New Policies Scenario 555 16.16 Refinery runs and demand in selected Asian countries and regions in the New Policies Scenario 556 16.17 Oil product trade balance in selected Asian regions in the New Policies Scenario 556 16.18 Allocation of Russian crude oil and condensate production in the New Policies Scenario 557 16.19 Main oil product yields in Russian refineries in the New Policies Scenario 558 16.20 Changes in refinery runs and exports in Iraq and the rest of the Middle East in the New Policies Scenario, 2012-2035 559 16.21 Middle East trade of selected oil products in the New Policies Scenario 559 16.22 Crude export and refinery runs in selected crude oil-exporting countries and regions in the New Policies Scenario 560 16.23 Net oil imports in selected regions in the New Policies Scenario 561 16.24 Net oil trade in North America in the New Policies Scenario 562 16.25 Combined crude oil trade balance of Middle East and Asia in the New Policies Scenario 563 16.26 Crude oil imports by region and source in the New Policies Scenario 564

List of tables Part A: GLOBAL ENERGY TRENDS Tables for Chapter 1: Scope and methodology

© OECD/IEA, 2013

1.1 Overview of key assumptions and energy prices in the New Policies Scenario 1.2 Real GDP growth assumptions by region 1.3 Population assumptions by region 1.4 Fossil fuel import prices by scenario 1.5 CO2 price assumptions in selected regions by scenario 1.6 Recent progress and key conditions for faster deployment of clean energy technologies Tables for Chapter 2: Global energy trends to 2035

2.1

World primary energy demand and energy-related CO2 emissions by scenario

Table of Contents

36 40 42 48 51 52 58

2.2 World primary energy demand by region in the New Policies Scenario 2.3 Number of people without access to modern energy services by region, 2011 2.4 Number of people without access to modern energy services by region in the New Policies Scenario, 2011 and 2030

69 89 91

Tables for Chapter 3: Natural gas market outlook

3.1 Natural gas demand and production by region and scenario 100 3.2 Natural gas demand by region in the New Policies Scenario 103 3.3 Remaining technically recoverable natural gas resources by type and region, end-2012 108 3.4 Natural gas production by region in the New Policies Scenario 109 3.5 Global production of unconventional gas in the New Policies Scenario 117 3.6 Net natural gas trade by region in the New Policies Scenario (pipeline and LNG) 124 3.7 Indicative range of cost estimates for conversion and inter-regional transportation for LNG, 2020 133 Tables for Chapter 4: Coal market outlook

4.1 4.2 4.3 4.4

Coal demand, production and trade by scenario Coal demand by region in the New Policies Scenario Coal production by region in the New Policies Scenario Inter-regional coal trade in the New Policies Scenario

141 145 149 150

Tables for Chapter 5: Power sector outlook

5.1 Electricity demand by region and scenario 5.2 World electricity demand by sector and generation in the New Policies Scenario 5.3 Electricity generation by source and scenario 5.4 Cumulative capacity retirements by region and source in the New Policies Scenario, 2013-2035 5.5 Cumulative gross capacity additions by region and source in the New Policies Scenario, 2013-2035 Tables for Chapter 6: Renewable energy outlook

6.1 World renewable energy use by type and scenario 6.2 Renewables-based electricity generation by region in the New Policies Scenario 6.3 Ethanol and biodiesel consumption in road transport by region in the New Policies Scenario

172 173 175 178 179 200 202 205

© OECD/IEA, 2013

Tables for Chapter 7: Energy efficiency outlook

7.1 Selected energy efficiency policies announced or introduced in 2012 and 2013 239 7.2 Key energy efficiency assumptions in major regions in the New Policies Scenarios 244 7.3 Savings in industrial energy demand and CO2 emissions from energy efficiency in the New Policies scenario 246

World Energy Outlook 2013

7.4 Savings in transport energy demand and CO2 emissions from energy efficiency in the New Policies Scenario 7.5 Savings in buildings energy demand and CO2 emissions from energy efficiency in the New Policies scenario Tables for Chapter 8: Energy and competitiveness

8.1 8.2 8.3 8.4

Share of industry in final energy use by fuel and region, 2011 Share of tax in industrial energy prices in selected countries, 2012 Indicators of significance of industry to the economy by subsector and region Typology of main energy-intensive industries

248 250 264 271 277 280

Part B: BRAZIL ENERGY OUTLOOK Tables for Chapter 9: The Brazilian energy sector today

9.1 GDP and population indicators and assumptions 9.2 Main policy assumptions for Brazil in the New Policies Scenario Tables for Chapter 10: Prospects for Brazil’s domestic energy consumption

10.1 10.2 10.3

Brazil total primary energy demand by fuel and scenario Brazil final energy consumption by sector in the New Policies Scenario Brazil oil demand by product in the New Policies Scenario

Tables for Chapter 11: Brazilian resources and supply potential

11.1 11.2 11.3

Brazil conventional oil resources by region Brazil gas resources by region Indicative oil development and production costs in selected regions

Tables for Chapter 12: Implications of Brazil’s energy development

12.1

Brazil supply-demand balance by fuel in the New Policies Scenario

323 326 331 344 355 364 365 366 405

Part C: OUTLOOK FOR OIL MARKETS Tables for Chapter 13: From oil resources to reserves

13.1 Remaining recoverable oil resources and proven reserves, end-2012 13.2 Comparative economics of LTO and EOR hypothetical projects 13.3 Major LTO resource-holders

© OECD/IEA, 2013

Tables for Chapter 14: Prospects for oil supply

14.1 Oil production and supply by source and scenario 14.2 Breakdown of the field database by field size (recoverable resources) and geographic location 14.3 Weighted average CADR to 2012 by decline phase 14.4 World oil production by type in the New Policies Scenario 14.5 Non-OPEC oil production in the New Policies Scenario

Table of Contents

423 444 449 458 462 464 471 481

14.6 OPEC oil production in the New Policies Scenario 14.7 Cumulative investment in upstream oil and gas supply by region in the New Policies Scenario, 2013-2035 14.8 Oil and gas industry investment by company Tables for Chapter 15: Prospects for oil demand

15.1 15.2 15.3 15.4 15.5 15.6

Oil and total liquids demand by scenario Oil demand by region in the New Policies Scenario New policies in 2012/2013 with a potential impact on oil demand Oil demand by sector in the New Policies Scenario Main sources and uses of oil products, 2012 World primary oil demand by product in the New Policies Scenario

484 496 498 503 505 510 511 529 529

Tables for Chapter 16: Implications for oil refining and trade

16.1 Global total demand for liquids, products and crude throughput in the New Policies Scenario 16.2 World refining capacity and refinery runs in the New Policies Scenario

537 545

List of boxes Part A: GLOBAL ENERGY TRENDS Boxes for Chapter 1: Scope and methodology

1.1 1.2 1.3

Recent key developments in energy and environmental policy Uncertainties around the economic outlook Deriving the fossil fuel prices used in WEO analysis

Boxes for Chapter 2: Global energy trends to 2035

2.1 2.2 2.3

Building on a new base Fuel shortages in Pakistan Smuggling as a possible driver of fossil-fuel subsidy reform

Boxes for Chapter 3: Natural gas market outlook

3.1 3.2 3.3 3.4 3.5 3.6 3.7 3.8

Wide variety in regional starting points for the gas outlook Natural gas liquids and upstream gas investment Levant gas on the rise High-Level Unconventional Gas Forum-towards global best practice Are methane hydrates the next revolution-in-waiting? How great should expectations be for North American LNG? Price differentials between regions in a “global” gas market Trinidad and Tobago: seeking a new foothold in a changing gas world

© OECD/IEA, 2013

Boxes for Chapter 4: Coal market outlook

4.1 4.2

A quick guide to the different types of coal Was 2012 an aberration or a harbinger of change in coal demand?

35 39 44 56 90 94 101 112 114 115 119 127 133 138 142 144

World Energy Outlook 2013

4.3 4.4

Steam coal trade thrives as demand stutters US President’s Climate Action Plan

Boxes for Chapter 5: Power sector outlook

5.1

Coal-to-gas switching in the power sector

Boxes for Chapter 6: Renewable energy outlook

6.1 Reducing the challenges posed by variable renewables 6.2 Variable renewables in the 450 Scenario 6.3 Multiple benefits of renewables

151 161 184 213 216 226

Boxes for Chapter 7: Energy efficiency outlook

7.1 The Efficient World Scenario – tackling competitiveness, energy security and climate change simultaneously 7.2 Energy efficiency, energy intensity and energy savings 7.3 Energy efficiency does deliver Boxes for Chapter 8: Energy and competitiveness

8.1 8.2 8.3 8.4 8.5  8.6

Defining “competitiveness” Effect of taxes and subsidies on competitiveness Reindustrialisation of the US economy: myth or reality? The remarkable renaissance of US petrochemicals Expensive energy adds to the steel woes of the European Union Energy competitiveness and the European Union

233 234 236 262 270 281 285 286 298

Part B: BRAZIL ENERGY OUTLOOK Boxes for Chapter 9: The Brazilian energy sector today

9.1 Electricity crisis in Brazil, 2001-2002 9.2 Bioenergy in Brazil: more than ethanol 9.3 Brazil’s upstream regulatory framework

Boxes for Chapter 10: Prospects for Brazil’s domestic energy consumption

10.1 10.2 10.3 10.4 10.5 10.6

End-user energy efficiency policies in Brazil What if hydropower falls short? Brazil’s high tolerance for variable renewables The upstream oil and gas industry – a major consumer of its own products Keeping Brazil cool A long and winding road to a competitive Brazilian gas market

© OECD/IEA, 2013

Boxes for Chapter 11: Brazilian resources and supply potential

11.1 11.2 11.3 11.4

Petrobras and global deepwater oil production Brazil and deepwater regulation after Macondo Could oil suffer from divided royalties? Unconventional gas outlook for Brazil

Table of Contents

311 313 317 334 336 343 348 353 357 368 371 376 382

11.5 11.6

A platform for Amazon hydropower development Prospects for a new generation of biofuels in Brazil

383 390

Boxes for Chapter 12: Implications of Brazil’s energy development

12.1

Choices for oil production beyond self-sufficiency

402

Part C: OUTLOOK FOR OIL MARKETS Boxes for Chapter 13: From oil resources to reserves

13.1 Light tight oil, conventional or unconventional? 13.2 Oil resources and reserves under different classification systems 13.3 Grouping oil and gas companies in the WEO-2013 analysis 13.4 The risk of “stranded assets” in the upstream oil sector 13.5 Recovery rates and the case for EOR 13.6 CO2 enhanced oil recovery for carbon capture and storage Boxes for Chapter 14: Prospects for oil supply

14.1 14.2 14.3 14.4 14.5

Why does production decline? Concepts used in the decline rate analysis Will LTO techniques improve recovery at conventional reservoirs? The rising overseas presence of Asian national oil companies Staffing the oil and gas business

Boxes for Chapter 15: Prospects for oil demand

15.1 15.2 15.3 15.4

Could the world ever fall out of love with the automobile? Of chickens, eggs, trucks and cars A guide to petrochemicals Is cheap coal the Chinese answer to cheap gas in the United States?

Boxes for Chapter 16: Implications for oil refining and trade

16.1

Is history repeating itself for European refining?

424 431 433 436 439 441 460 461 475 494 500 514 518 524 527 551

List of spotlights

© OECD/IEA, 2013

Part A: GLOBAL ENERGY TRENDS How does the IEA model future energy trends? Redrawing the energy-climate map Is China’s coal demand set to peak soon? Is residential solar PV already competitive? Policies for energy efficiency in buildings in China What role does energy play in Korea’s industrial success?

41 80 146 218 252 266

World Energy Outlook 2013

© OECD/IEA, 2013

Part B: BRAZIL ENERGY OUTLOOK Are high energy prices hurting Brazil’s industrial competitiveness? Local content in Brazil; short-term costs, long-term value? How might climate change affect Brazil’s energy sector?

346 373 412

Part C: OUTLOOK FOR OIL MARKETS Has the rise of LTO resolved the debate about peak oil? What does the rise of light tight oil mean for decline rates?

447 467

Table of Contents

© OECD/IEA, 2013

džĞĐƵƟǀĞ^ƵŵŵĂƌLJ KƌŝĞŶƚĂƟŽŶĨŽƌĂĨĂƐƚͲĐŚĂŶŐŝŶŐĞŶĞƌŐLJǁŽƌůĚ Many of tŚe lonŐ-Śeld tenets of tŚe enerŐy sector are beinŐ reǁriƩen. Major importers are becoming exporters, while countries long-deĮned as major energy exporters are also becoming leading centres of global demand growth. The right combinaƟon of policies and technologies is proving that the links between economic growth, energy demand and energy-related CO2 emissions can be weakened. The rise of unconvenƟonal oil and gas and of renewables is transforming our understanding of the distribuƟon of the world’s energy resources. Awareness of the dynamics underpinning energy markets is essenƟal for decisionmakers aƩempƟng to reconcile economic, energy and environmental objecƟves. Those that anƟcipate global energy developments successfully can derive an advantage, while those that fail to do so risk making poor policy and investment decisions. This ediƟon of the World Energy Outlookථ(WEO-2013) examines the implicaƟons of diīerent sets of choices for energy and climate trends to 2035, providing insights along the way that can help policymakers, industry and other stakeholders Įnd their way in a fast-changing energy world.

© OECD/IEA, 2013

The centre of gravity of energy demand is switching decisively to the emerging economies, ƉarƟcularly China, India and the Middle East, which drive global energy use one-third higher. In the New Policies Scenario, the central scenario of WEO-2013, China dominates the picture within Asia, before India takes over from 2020 as the principal engine of growth. Southeast Asia likewise emerges as an expanding demand centre (a development covered in detail in the WEO Special Report: Southeast Asia Energy Outlook, published in October 2013). China is about to become the largest oil-imporƟng country and India becomes the largest importer of coal by the early 2020s. The United States moves steadily towards meeƟng all of its energy needs from domesƟc resources by 2035. Together, these changes represent a re-orientaƟon of energy trade from the AtlanƟc basin to the AsiaPaciĮc region. High oil prices, persistent diīerences in gas and electricity prices between regions and rising energy import bills in many countries focus aƩenƟon on the relaƟonship between energy and the broader economy. The links between energy and development are illustrated clearly in Africa, where, despite a wealth of resources, energy use per capita is less than one-third of the global average in 2035. Africa today is home to nearly half of the 1.3ථbillion people in the world without access to electricity and one-quarter of the 2.6ථbillion people relying on the tradiƟonal use of biomass for cooking. Globally, fossil fuels conƟnue to meet a dominant share of global energy demand, with implicaƟons for the links between energy, the environment and climate change. As the source of two-thirds of global greenhouse-gas emissions, the energy sector will be Ɖivotal in determining whether or not climate change goals are achieved. Although some carbon abatement schemes have come under pressure, iniƟaƟves such as the President’s Climate AcƟon Plan in the United States, the Chinese plan to limit the share of coal in the domesƟc energy mix, the European debate on 2030 energy and climate targets and Japan’s

Executive Summary

23

discussions on a new energy plan all have the potenƟal to limit the growth in energyrelated CO2 emissions. In our central scenario, taking into account the impact of measures already announced by governments to improve energy eĸciency, support renewables, reduce fossil-fuel subsidies and, in some cases, to put a price on carbon, energy-related CO2 emissions sƟll rise by 20й to 2035. This leaves the world on a trajectory consistent with a long-term average temperature increase of 3.6ථΣC, far above the internaƟonally agreed 2ථΣC target.

tŚŽŚĂƐƚŚĞĞŶĞƌŐLJƚŽĐŽŵƉĞƚĞ͍ Large diīerences in regional energy Ɖrices have sƉarŬed a debate about the role of energy in unleashing or frustraƟng economic growth. Brent crude oil has averaged Ψ110ථperථbarrel in real terms since 2011, a sustained period of high oil prices that is without parallel in oil market history. But unlike crude oil prices, which are relaƟvely uniform worldwide, prices of other fuels have been subject to signiĮcant regional variaƟons. Although gas price diīerenƟals have come down from the extraordinary levels seen in mid-2012, natural gas in the United States sƟll trades at one-third of import prices to Europe and one-ĮŌh of those to Japan. Electricity prices also vary, with average Japanese or European industrial consumers paying more than twice as much for power as their counterparts in the United States, and even Chinese industry paying almost double the US level. In most sectors, in most countries, energy is a relaƟvely minor part of the calculaƟon of compeƟƟveness. But energy costs can be of crucial importance to energy-intensive industries, such as chemicals, aluminium, cement, iron and steel, paper, glass and oil reĮning, parƟcularly where the resulƟng goods are traded internaƟonally. Energy-intensive sectors worldwide account for around one-ĮŌh of industrial value added, one-quarter of industrial employment and 70й of industrial energy use. Energy Ɖrice variaƟons are set to aīect industrial comƉeƟƟveness, inŇuencing investment decisions and comƉany strategies. While regional diīerences in natural gas prices narrow in our central scenario, they nonetheless remain large through to 2035 and, in most cases, electricity price diīerenƟals persist. In many emerging economies, parƟcularly in Asia, strong growth in domesƟc demand for energy-intensive goods supports a swiŌ rise in their producƟon (accompanied by export expansion). But relaƟve energy costs play a more decisive role in shaping developments elsewhere. The United States sees a slight increase in its share of global exports of energy-intensive goods, providing the clearest indicaƟon of the link between relaƟvely low energy prices and the industrial outlook. By contrast, the European Union and Japan both see a strong decline in their export shares ʹ a combined loss of around one-third of their current share.

© OECD/IEA, 2013

^ĞĂƌĐŚŝŶŐĨŽƌĂŶĞŶĞƌŐLJƐƚƚŽƚŚĞĞĐŽŶŽŵLJ Countries can reduce the imƉact of high Ɖrices by ƉromoƟng more eĸcient, comƉeƟƟve and interconnected energy marŬets. Cost diīerenƟals between regional gas markets could be narrowed further by more rapid movement towards a global gas market. As we examine in a Gas Price Convergence Case, this would require a loosening of the current rigidity of liqueĮed natural gas (LNG) contracƟng structures and oil-indexed pricing mechanisms, 24

World Energy Outlook 2013

spurred by accelerated gas market reforms in the Asia-PaciĮc region and LNG exports from North America (and an easing of costs for LNG liquefacƟon and shipping). There is also potenƟal in some regions, notably China, parts of LaƟn America and even parts of Europe, to replicate at smaller scale the US success in developing its unconvenƟonal gas resources, though uncertainty remains over the quality of the resources, the costs of their producƟon and, in some countries, public acceptance for their development. A renewed focus on energy eĸciency is taŬing hold and is set to deliver beneĮts that extend well beyond imƉrovements in comƉeƟƟveness. Notable policies introduced over the past year include measures targeƟng eĸciency improvements in buildings in Europe and Japan, in motor vehicles in North America and in air condiƟoners in parts of the Middle East, and energy pricing reforms in China and India. As well as bringing down costs for industry, eĸciency measures miƟgate the impact of energy prices on household budgets (the share of energy in household spending has reached very high levels in the European Union) and on import bills (the share of energy imports in Japan’s GDP has risen sharply). But the potenƟal for energy eĸciency is sƟll far from exhausted: two-thirds of the economic potenƟal of energy eĸciency is set to remain untapped in our central scenario. AcƟon is needed to break down the various barriers to investment in energy eĸciency. This includes phasing out fossil-fuel subsidies, which we esƟmate rose to Ψ544ථbillion worldwide in 2012. Enhancing energy comƉeƟƟveness does not mean diminishing eīorts to tacŬle climate change. The WEO Special Report: Redrawing the Energy-Climate Map, published in June 2013 idenƟĮed four pragmaƟc measures ʹ improving eĸciency, limiƟng the construcƟon and use of the least-eĸcient coal-Įred power plants, minimising methane emissions in upstream oil and gas, and reforming fossil-fuel subsidies ʹ that could halt the increase in emissions by 2020 without harming economic growth. This package of measures would complement the developments already envisaged in our central scenario, notably the rise in deployment of renewable energy technologies. Governments need, though, to be aƩenƟve to the design of their subsidies to renewables, which surpassed Ψ100 billion in 2012 and expand to Ψ220ථbillion in 2035. As renewables become increasingly compeƟƟve on their own merits, it is important that subsidy schemes allow for the mulƟple beneĮts of low-carbon energy sources without placing excessive burdens on those that cover the addiƟonal costs. A carefully conceived internaƟonal climate change agreement can help to ensure that the energy-intensive industries in countries that act decisively to limit emissions do not face unequal compeƟƟon from countries that do not.

© OECD/IEA, 2013

>ŝŐŚƚƟŐŚƚŽŝůƐŚĂŬĞƐƚŚĞŶĞdžƚƚĞŶLJĞĂƌƐ͕ďƵƚůĞĂǀĞƐƚŚĞůŽŶŐĞƌƚĞƌŵƵŶƐƟƌƌĞĚ The caƉacity of technologies to unlocŬ new tyƉes of resources, such as light Ɵght oil ;LTKͿ and ultra-deeƉwater Įelds, and to imƉrove recovery rates in exisƟng Įelds is Ɖushing uƉ esƟmates of the amount of oil that remains to be Ɖroduced. But this does not mean that the world is on the cusp of a new era of oil abundance. An oil price that rises steadily to Ψ128 per barrel (in year-2012 dollars) in 2035 supports the development of these new resources, though no country replicates the level of success with LTO that is making the Executive Summary

25

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18

United States the largest global oil producer. The rise of unconvenƟonal oil (including LTO) and natural gas liquids meets the growing gap between global oil demand, which rises by 14ථmbͬd to reach 101ථmbͬd in 2035, and producƟon of convenƟonal crude oil, which falls back slightly to 65ථmbͬd. The Middle East, the only large source of low-cost oil, remains at the centre of the longer-term oil outlooŬ. The role of OPEC countries in quenching the world’s thirst for oil is reduced temporarily over the next ten years by rising output from the United States, from oil sands in Canada, from deepwater producƟon in Brazil and from natural gas liquids from all over the world. But, by the mid-2020s, non-OPEC producƟon starts to fall back and countries in the Middle East provide most of the increase in global supply. Overall, naƟonal oil companies and their host governments control some 80й of the world’s proven-plusprobable oil reserves. The need to comƉensate for declining outƉut from exisƟng oil Įelds is the major driver for uƉstream oil investment to ϮϬϯ5. Our analysis of more than 1ථ600 Įelds conĮrms that, once producƟon has peaked, an average convenƟonal Įeld can expect to see annual declines in output of around 6й per year. While this Įgure varies according to the type of Įeld, the implicaƟon is that convenƟonal crude output from exisƟng Įelds is set to fall by more than 40ථmbͬd by 2035. Among the other sources of oil, most unconvenƟonal plays are heavily dependent on conƟnuous drilling to prevent rapid Įeld-level declines. Of the 790ථbillion barrels of total producƟon required to meet our projecƟons for demand to 2035, more than half is needed just to oīset declining producƟon. Demand for mobility and for Ɖetrochemicals ŬeeƉs oil use on an uƉward trend to ϮϬϯ5, although the Ɖace of growth slows. The decline in oil use in OECD countries accelerates. China overtakes the United States as the largest oil-consuming country and Middle East oil consumpƟon overtakes that of the European Union, both around 2030. The shiŌing geography of demand is further underlined by India becoming the largest single source of global oil demand growth aŌer 2020. Oil consumpƟon is concentrated in just two sectors by 2035: transport and petrochemicals. Transport oil demand rises by 25й to reach 59ථmbͬd, with one-third of the increase going to fuel road freight in Asia. In petrochemicals, the Middle East, China and North America help push up global oil use for feedstocks to 14ථmbͬd. High prices encourage eĸciency improvements and undercut the posiƟon of oil wherever alternaƟves are readily available, with biofuels and natural gas gaining some ground as transport fuels.

© OECD/IEA, 2013

dŚĞŐƌĞĂƚŵŝŐƌĂƟŽŶŝŶŽŝůƌĞĮŶŝŶŐĂŶĚƚƌĂĚĞ Major changes in the comƉosiƟon of oil suƉƉly and demand confront the world͛s reĮners with an ever-more comƉlex set of challenges, and not all of them are well-eƋuiƉƉed to survive. Rising output of natural gas liquids, biofuels and coal-ථorථgas-to-liquids technologies means that a larger share of liquid fuels reaches consumers without having to pass through the reĮnery system. ReĮners nonetheless need to invest to meet a surge of more than 5ථmbͬd in demand for diesel that is almost triple the increase in gasoline 26

World Energy Outlook 2013

use. The shiŌ in the balance of oil consumpƟon towards Asia and the Middle East sees a conƟnued build-up of reĮning capacity in these regions͖ but, in many OECD countries, declining demand and compeƟƟon in product export markets intensify pressure to shut capacity. Over the period to 2035, we esƟmate that nearly 10ථmbͬd of global reĮnery capacity is at risk, with reĮneries in OECD countries, and Europe in parƟcular, among the most vulnerable. The new geograƉhy of demand and suƉƉly means a re-ordering of global oil trade Ňows towards Asian marŬets, with imƉlicaƟons for co-oƉeraƟve eīorts to ensure oil security. The net North American requirement for crude imports all but disappears by 2035 and the region becomes a larger exporter of oil products. Asia becomes the unrivalled centre of global oil trade as the region draws in ʹ via a limited number of strategic transport routes ʹ a rising share of the available crude oil. Deliveries to Asia come not only from the Middle East (where total crude exports start to fall short of Asian import requirements) but also from Russia, the Caspian, Africa, LaƟn America and Canada. New export-oriented reĮnery capacity in the Middle East raises the possibility that oil products, rather than crude, take a larger share of global trade, but much of this new capacity eventually serves to cater to increasing demand from within the region itself.

dŚĞƉŽǁĞƌƐĞĐƚŽƌĂĚũƵƐƚƐƚŽĂŶĞǁůŝĨĞǁŝƚŚǁŝŶĚĂŶĚƐŽůĂƌ Renewables account for nearly half of the increase in global Ɖower generaƟon to ϮϬϯ5, with variable sources ʹ wind and solar Ɖhotovoltaics ʹ maŬing uƉ 45й of the exƉansion in renewables. China sees the biggest absolute increase in generaƟon from renewable sources, more than the increase in the European Union, the United States and Japan combined. In some markets, the rising share of variable renewables creates challenges in the power sector, raising fundamental quesƟons about current market design and its ability to ensure adequate investment and long-term reliability of supply. The increase in generaƟon from renewables takes its share in the global power mix above 30й, drawing ahead of natural gas in the next few years and all but reaching coal as the leading fuel for power generaƟon in 2035. The current rate of construcƟon of nuclear power plants has been slowed by reviews of safety regulaƟons, but output from nuclear eventually increases by two-thirds, led by China,
© OECD/IEA, 2013

ĐŽŶŽŵŝĐƐĂŶĚƉŽůŝĐŝĞƐ͕ŝŶĚŝīĞƌĞŶƚĚŽƐĞƐ͕ĂƌĞŬĞLJƚŽƚŚĞŽƵƚůŽŽŬĨŽƌĐŽĂůĂŶĚŐĂƐ Coal remains a cheaƉer oƉƟon than gas for generaƟng electricity in many regions, but Ɖolicy intervenƟons to imƉrove eĸciency, curtail local air ƉolluƟon and miƟgate climate change will be criƟcal in determining its longer-term ƉrosƉects. Policy choices in China, which has outlined plans to cap the share of coal in total energy use, will be parƟcularly important as China now uses as much coal as the rest of the world combined. In our central scenario, global coal demand increases by 17й to 2035, with two-thirds of the increase Executive Summary

27

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18

occurring by 2020. Coal use declines in OECD countries. By contrast, coal demand expands by one-third in non-OECD countries ʹ predominantly in India, China and Southeast Asia ʹ despite China reaching a plateau around 2025. India, Indonesia and China account for 90й of the growth in coal producƟon. Export demand makes Australia the only OECD country to register substanƟal growth in output. MarŬet condiƟons vary striŬingly in diīerent regions of the world, but the Ňexibility and environmental beneĮts of natural gas comƉared with other fossil fuels Ɖut it in a ƉosiƟon to ƉrosƉer over the longer term. Growth is strongest in emerging markets, notably China, where gas use quadruples by 2035, and in the Middle East. But in the European Union, gas remains squeezed between a growing share of renewables and a weak compeƟƟve posiƟon versus coal in power generaƟon, and consumpƟon struggles to return to 2010 levels. North America conƟnues to beneĮt from ample producƟon of unconvenƟonal gas, with a small but signiĮcant share of this gas Įnding its way to other markets as LNG, contribuƟng ʹ alongside other convenƟonal and unconvenƟonal developments in East Africa, China, Australia and elsewhere ʹ to more diversity in global gas supply. New connecƟons between markets act as a catalyst for changes in the way that gas is priced, including more widespread adopƟon of hub-based pricing.

ƌĂnjŝůŝƐĂƚƚŚĞůĞĂĚŝŶŐĞĚŐĞŽĨĚĞĞƉǁĂƚĞƌĂŶĚůŽǁͲĐĂƌďŽŶĚĞǀĞůŽƉŵĞŶƚ

© OECD/IEA, 2013

Brazil, the sƉecial focus country in this year͛s KƵƚůŽŽŬ, is set to become a major exƉorter of oil and a leading global energy Ɖroducer. Based mainly on a series of recent oīshore discoveries, Brazil’s oil producƟon triples to reach 6ථmbͬd in 2035, accounƟng for onethird of the net growth in global oil producƟon and making Brazil the world’s sixth-largest producer. Natural gas producƟon grows more than Įve-fold, enough to cover all of the country’s domesƟc needs by 2030, even as these expand signiĮcantly. The increase in oil and gas producƟon is dependent on highly complex and capital-intensive deepwater developments, requiring levels of upstream investment beyond those of either the Middle East or Russia. A large share of this will need to come from Petrobras, the naƟonal oil company, whose mandated role in developing strategic Įelds places heavy weight on its ability to deploy resources eīecƟvely across a huge and varied investment programme. Commitments made to source goods and services locally within Brazil add tension to a Ɵghtly stretched supply chain. Brazil͛s abundant and diverse energy resources underƉin an ϴϬй increase in its energy use, including the achievement of universal access to electricity. Rising consumpƟon is driven by the energy needs of an expanding middle class, resulƟng in strong growth in demand for transport fuels and a doubling of electricity consumpƟon. MeeƟng this demand requires substanƟal and Ɵmely investment throughout the energy system ʹ Ψ90ථbillion per year on average. The system of aucƟons for new electricity generaƟon and transmission capacity will be vital in bringing new capital to the power sector and in reducing pressure on end-user prices. The development of a well-funcƟoning gas market, aƩracƟve to new entrants, can likewise help spur investment and improve the compeƟƟve posiƟon of Brazilian industry. A stronger policy focus on energy eĸciency would ease potenƟal strains on a rapidly growing energy system. 28

World Energy Outlook 2013

Brazil͛s energy sector remains one of the least carbon-intensive in the world, desƉite greater availability and use of fossil fuels. Brazil is already a world-leader in renewable energy and is set to almost double its output from renewables by 2035, maintaining their 43й share of the domesƟc energy mix. Hydropower remains the backbone of the power sector. zet reliance on hydropower declines, in part because of the remoteness and environmental sensiƟvity of a large part of the remaining resource, much of which is in the Amazon region. Among the fuels with a rising share in the power mix, onshore wind power, which is already proving to be compeƟƟve, natural gas and electricity generated from bioenergy take the lead. In the transport sector, Brazil is already the world’s secondlargest producer of biofuels and its producƟon, mainly as sugarcane ethanol, more than triples. Suitable culƟvaƟon areas are more than suĸcient to accommodate this increase without encroaching on environmentally sensiƟve areas. By 2035, Brazilian biofuels meet almost one-third of domesƟc demand for road-transport fuel and its net exports account for about 40й of world biofuels trade.

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

© OECD/IEA, 2013

17 18 Executive Summary

29

© OECD/IEA, 2013

PART A GLOBAL ENERGY TRENDS

PREFACE

Part A of this WEO (Chapters 1-8) presents energy projecƟons to 2035. It covers the prospects for all energy sources, regions and sectors and an assessment of the impact of energy use on climate change. Three scenarios are presented ʹ the New Policies Scenario, the Current Policies Scenario and the 450 Scenario ʹ together with several special cases. Chapter 1 deĮnes the scenarios and sets out the various inputs and modelling assumpƟons uƟlised in the analysis. Chapter 2 summarises the results of the projecƟons for global energy in aggregate and draws out the implicaƟons for energy security, environmental protecƟon and economic development. The chapter also provides special features on Southeast Asia’s emergence as a key player in the global energy system, achieving universal energy access and developments in subsidies to fossil fuels and renewables. Chapters 3-6 analyse the outlook for natural gas, coal, electricity and renewables. Chapter 7 covers the current status and future prospects for energy eĸciency, which is treated for the very Įrst Ɵme in the same way as the convenƟonal energy sources, with its own standalone chapter.

© OECD/IEA, 2013

Chapter 8 examines energy and compeƟƟveness, assessing what major dispariƟes in regional energy prices might mean for economies, parƟcularly their energy-intensive industries.

© OECD/IEA, 2013

Chapter 1 Scope and methodology What underlies the analysis? Highlights

x The New Policies Scenario – the central scenario in WEO-2013 – analyses the evoluƟon of energy markets based on the conƟnuaƟon of exisƟng policies and measures as well as cauƟous implementaƟon of policies that have been announced by governments but are yet to be given eīect. The Current Policies Scenario takes account only of policies already enacted as of mid-2013. The 450 Scenario shows what it takes to set the energy system on track to have a 50й chance of keeping to 2ථΣC the long-term increase in average global temperature.

x More than Įve years aŌer the worst recession since the 1930s began in 2008, the economic recovery conƟnues to be fragile and uneven. We assume world GDP grows at an average rate of 3.6й per year through to 2035. This equates to a more than doubling in the size of the global economy. Developing Asia accounts for over half of the increase in economic acƟvity. China’s income per capita grows by around threeand-a-half Ɵmes, overtaking that of the Middle East.

x Demographic factors will conƟnue to drive changes in the energy mix. The world populaƟon is set to rise from 7.0ථbillion in 2011 to 8.7ථbillion in 2035, led by Africa and India. China’s populaƟon changes liƩle and by around 2025 India becomes the world’s most populous country. Most OECD countries see small changes in populaƟon, with the notable excepƟon of the United States, which sees an increase of about 60ථmillion people. Global populaƟon growth is concentrated enƟrely in urban areas.

x The world is experiencing a period of historically high oil prices. Brent crude oil has averaged over Ψ110ͬbarrel in real terms since 2011, a sustained period of high oil prices that is without parallel in oil market history. In the New Policies Scenario, oil prices reach Ψ113ͬbarrel in 2020 and Ψ128ͬbarrel in 2035. Big diīerences remain between gas prices in regional markets, despite some convergence. Coal prices remain much lower than oil and gas prices in energy equivalent terms. The share of global CO2 emissions subject to a CO2 price rises from 8й today to one-third in 2035.

x Energy technologies that are already in use or are approaching commercialisaƟon

© OECD/IEA, 2013

are assumed to achieve ongoing cost reducƟons as a result of increased learning and deployment. Although there are excepƟons that create some basis for opƟmism, recent progress in deploying clean energy technologies has not matched policy expectaƟons and, in many cases, their future uptake hinges on dedicated policy support andͬor subsidies.

Chapter 1 | Scope and methodology

33

Scope of report This ediƟon of the World Energy Outlook (WEO) presents an assessment of the prospects for global energy markets in the period to 2035 and draws out the implicaƟons for energy security, environmental protecƟon and economic development. The objecƟve is to provide policymakers, industry and the general public in countries all over the world with the data, analysis and insights needed to make judgements about our energy future, as a basis for sound energy decisionmaking. Part A of the report is built around projecƟons of energy demand and supply through to 2035. The three main scenarios – the New Policies Scenario, the Current Policies Scenario and the 450ථScenario – are underpinned by assumpƟons about economic and populaƟon growth, and about energy and climate policies and technology deployment. Energy prices are derived from a modelling process. Our analysis takes into account all of the historical energy data available to the IEA at the Ɵme of wriƟng, as well as more recent preliminary data from a wide variety of sources. For the Įrst Ɵme, energy eĸciency – a major factor in the global energy balance – is treated in much the same way as the convenƟonal fuels, its prospects being presented in a dedicated chapter that builds on the special focus on energy eĸciency included in WEO-2012 (IEA, 2012a). Part A also includes, in Chapterථ2, an update on three key areas of criƟcal importance to energy and climate trends: (i) achieving universal energy access͖ (ii) developments in subsidies to fossil fuels and renewables͖ and (iii) the impact of energy use on climate change. Prospects for unconvenƟonal gas producƟon, including the uptake of the IEA’s ͞Golden Rules͟ to address the associated environmental and social impacts, are included in Chapterථ3 (IEA, 2012b). Part A ends with an examinaƟon of energy and compeƟƟveness, assessing what major dispariƟes in regional energy prices might mean for consumers and the economy at large, and oīering insights into the policies that might be pursued to improve energy compeƟƟveness. Consistent with recent pracƟce, WEO-2013 includes a parƟcular focus on one country and on one energy source. The country of focus is Brazil, presented in Part B. We analyse how Brazil’s vast and diverse energy resource base – from renewables to new oīshore discoveries – can meet its growing domesƟc needs and help it to open up new export markets. The highlighted energy source is oil, presented in PartථC. We provide a fresh look at the oil resource base, the economics and decline rates of diīerent types of oil producƟon, the outlook for light Ɵght oil in North America and beyond, oil demand by product and the prospects for the reĮning sector.

© OECD/IEA, 2013

Introducing the scenarios Throughout the last year, signiĮcant new energy and environmental policies have been adopted in many parts of the world. A number of naƟonal energy strategy reviews have also been launched, which can be expected to lead to further new policy announcements in the near future. And some important progress has been made in bilateral and mulƟlateral energy co-operaƟon. These developments guide the diīerent policy assumpƟons adopted in the three scenarios (Box 1.1). 34

World Energy Outlook 2013 | Global Energy Trends

Box 1.1 ‫ ٲ‬Recent key developments in energy and environmental policy

1

Important policy developments in 2013 have been taken into account, to varying degrees, in the three scenarios presented in WEO-2013. Early in the year, the United States extended various tax credits for renewable energy, energy eĸciency and alternaƟve fuel vehicles. The US administraƟon also announced a major climate acƟon plan that seeks to introduce (i) new standards for power plants͖ (ii) more funding and incenƟves for energy eĸciency and renewables͖ (iii) preparaƟons to safeguard the country from the impacts of climate change͖ and (iv) steps to provide more global leadership to reduce carbon emissions. Canada adopted new fuel-economy standards for cars and light trucks, to take eīect in 2017 with targets for 2025 (mirroring the US Corporate Average Fuel Economy standards approved in 2012), and regulaƟons to improve the fuel eĸciency of new heavy-duty vehicles. Japan widened the scope of its Top-Runner Program to include building materials and released a new economic growth strategy that includes a call to restart the country’s nuclear reactors, most of which have lain idle since the Fukushima Daiichi accident (subject to meeƟng new safety requirements). Italy introduced a new energy plan, which calls for further development of renewable energy as well as oil and gas. China announced plans to reduce the share of coal in total primary energy demand to 65й by 2017 and to speed up the introducƟon of new vehicle emissions standards. India mandated a 5й ethanol blend in gasoline and announced a target to expand power generaƟon from renewables.

© OECD/IEA, 2013

In addiƟon to these new measures and targets, many important reviews have been underway, some of which could lead to new policies. Japan is working on a new energy plan to be released in late 2013. The European Commission has been consulƟng on a 2030 framework for climate and energy policies, including the nature of any targets that may be set. Germany has an ongoing debate on its Energiewende (energy transiƟon), which is aimed at ambiƟous decarbonisaƟon of its energy system. France has been holding a naƟonal debate on energy (Débat sur la TransiƟon EnergéƟque) in advance of a new energy policy bill. India’s Planning Commission consƟtuted an expert group to propose a low-carbon strategy for growth. Brazil is updaƟng its long-term energy strategy to 2050 (Plano Nacional de Energia 2050), while Saudi Arabia is developing plans to diversify its energy mix and free-up more oil for export. Important developments have also occurred in bilateral and mulƟlateral energy cooperaƟon. These include the ͞Power Africa͟ partnership between the United States and governments and companies in sub-Saharan Africa, which was launched mid-year with the aim of doubling access to electricity within the region. The United States and China also agreed on a (non-binding) plan to cut their carbon emissions from heavyduty vehicles and coal-Įred power plants. And on the road to the reform of fossil-fuel subsidies – an issue on the agenda of both the G-20 and APEC – a growing number of countries, including China, India and Indonesia, have introduced major energy pricing reforms (see fossil-fuel subsidies secƟon in Chapterථ2). In terms of schemes that place a price on carbon, some new iniƟaƟves have been introduced, but others have come under challenge (see carbon markets secƟon towards the end of this chapter). Chapter 1 | Scope and methodology

35

2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18

The New Policies Scenario is the central scenario of this Outlook. In addiƟon to incorporaƟng the policies and measures that aīect energy markets and that had been adopted as of mid2013, it also takes account of other relevant commitments that have been announced, even when the precise implementaƟon measures have yet to be fully deĮned (Tableථ1.1). These commitments include programmes to support renewable energy and improve energy eĸciency, iniƟaƟves to promote alternaƟve fuels and vehicles, carbon pricing and policies related to the expansion or phase-out of nuclear energy, and iniƟaƟves taken by G-20 and Asia-PaciĮc Economic CooperaƟon (APEC) economies to reform fossil-fuel subsidies. We take a cauƟous view as to the extent to which these commitments will be implemented, as there are insƟtuƟonal, poliƟcal and economic circumstances in all regions that could stand in the way. Details of the key policy targets and measures taken into account in the New Policies Scenario (as well as in the other two scenarios presented in WEO-2013) are set out in AnnexථB. Table 1.1 ‫ ٲ‬Overview of key assumptions and energy prices in the New Policies Scenario Factor

AssumpƟons

Policies

ConƟnuaƟon of policies that had been legally enacted as of mid-2013 plus cauƟous implementaƟon of announced commitments and plans.

GDP growth

Global GDP increases at an average rate of 3.6й per year over 2011-2035 (based on GDP expressed in year-2012 dollars in purchasing power parity terms).

PopulaƟon growth

World populaƟon rises at an average rate of 0.9й per year, to 8.7ථbillion in 2035. The proporƟon of people living in urban areas rises from 52й in 2011 to 62й in 2035.

Energy pricing

Average IEA crude oil import price reaches Ψ128ͬbarrel (in year-2012 dollars) in 2035. A degree of convergence in natural gas prices occurs between the three major regional markets of North America, Asia-PaciĮc and Europe. Coal prices remain much lower than oil and gas prices on an energy-equivalent basis.

© OECD/IEA, 2013

Fossil-fuel subsidies Phased out in all net-imporƟng regions within ten years (at the latest) and in netexporƟng regions where speciĮc legislaƟon has already been adopted. CO2 pricing

New schemes that put a price on carbon are gradually introduced, with price levels gradually increasing.

Technology

Energy technologies – both on the demand and supply sides – that are in use today or are approaching the commercialisaƟon phase achieve ongoing cost reducƟons.

We also present the Current Policies Scenario, which takes into account only those policies and measures aīecƟng energy markets that were formally enacted as of mid-2013. In other words, it describes a future in which governments do not implement any recent commitments that have yet to be backed-up by legislaƟon or introduce other new policies bearing on the energy sector. The scenario is designed to provide a baseline picture of how global energy markets would evolve if established trends in energy demand and supply conƟnue unabated. It both illustrates the consequences of inacƟon and makes it possible to evaluate the potenƟal eīecƟveness of recent developments in energy and climate policy.

36

World Energy Outlook 2013 | Global Energy Trends

The ϰϱϬ෴^ĐĞŶĂƌŝŽ shows what is needed to set the global energy sector on a course compaƟble with a near 50й chance of limiƟng the long-term increase in the average global temperature to two degrees Celsius (2ථΣC). This scenario leads to a peak in the concentraƟon of greenhouse gases in the atmosphere around the middle of this century, at a level above 450ථparts per millionථ(ppm), but not so high as to be likely to precipitate changes that make the 2ථΣC objecƟve unaƩainable. The concentraƟon of greenhouse gases stabilises aŌer 2100 at around 450ථppm. For the period to 2020, policy acƟon aiming at fully implemenƟng the commitments under the Cancun Agreements, which were made at the 2010 United NaƟons Climate Change Conference in Mexico, is assumed to be undertaken (in the New Policies Scenario these commitments are only partly implemented). AŌer 2020, OECD countries and other major economies are assumed to implement emissions reducƟon measures that, collecƟvely, ensure a trajectory consistent with the target. From 2020, OECD countries are assumed to mobilise Ψ100ථbillion in annual Įnancing from a variety of sources for abatement measures in non-OECD countries. The 450ථScenario is not given the same coverage in WEO-2013 as in previous ediƟons as the speciĮc short-term opportuniƟes for acƟon in the energy sector to miƟgate climate change and their potenƟal results were covered in detail in Redrawing the Energy-Climate Map, a WEO special report that was released in June 2013 (IEA, 2013b). The results of the 450ථScenario are, however, included in many of the tables and Įgures in this report.

3 4 5 6 7 8

10

Economic growth More than Įve years aŌer the worst economic recession since the 1930s began in 2008, posiƟve signs of recovery in some economies cannot hide the fact that, overall, the recovery remains fragile and downside risks remain. It has been characterised as a two-speed recovery. Developing economies have been growing at much faster rates than advanced economies and make up the Įrst group (though many of them, including China, Russia, Brazil, India and in Southeast Asia have, more recently, been showing signs of slowing momentum ΀Box 1.2΁). Advanced economies make up the second group, though here too, there are divergences. Growth in the United States, Canada, Australia, New ealand,
2

9

Building blocks of the scenarios

1.ഩ The average rate of improvement, however, was much lower in 2000-2011 than in 1980-2000 (and energy intensity actually increased in 2009 and 2010) due to a shift in the balance of global economic activity to countries in developing Asia which have relatively high energy intensities (see Chapter 7).

Chapter 1 | Scope and methodology

1

37

11 12 13 14 15 16 17 18

has been parƟcularly evident in the OECD, the two sƟll remain closely Ɵed (Figureථ1.1). It follows that the projecƟons in this Outlook are highly sensiƟve to assumpƟons about the rates and paƩerns of GDP growth.

50

10 000

Mtoe

Trillion dollars (2012)

Figure 1.1 ‫ ٲ‬Primary energy demand and GDP GDP: OECD

40

8 000

Non-OECD

30

6 000

TPED (right axis): OECD Non-OECD

20

4 000

10

2 000

1971

1980

1990

2000

2012

Notes: Calculated on the basis of GDP in year-2012 dollars expressed in real purchasing power parity terms. TPED с total primary energy demand.

© OECD/IEA, 2013

In each of the scenarios in this Outlook, world GDP (expressed in real purchasing power parity ΀PPP΁ terms) is assumed to grow at an average annual rate of 3.6й between 2011 and 2035 (Tableථ1.2).2 This means that the global economy more than doubles in size over the period. Although this is just marginally faster overall than what was assumed in last year’s Outlook, there have been more signiĮcant revisions in some regions. These include downward revisions in the period to 2020 for India, Brazil, Russia and the European Union and upward revisions in the same period for the United States, among others. For the medium term, our GDP growth assumpƟons have been based primarily on InternaƟonal Monetary Fund (IMF) forecasts, with some adjustments to reŇect informaƟon available from regional, naƟonal and other sources. The latest IMF forecasts are for global economic growth of 2.9й in 2013 and 3.6й in 2014, before acceleraƟng to 4.1й annually in 2018 (IMF, 2013). Longer-term GDP assumpƟons are based on our assessment of prospects for growth in labour supply and improvements in producƟvity, supplemented by projecƟons made by various economic forecasƟng bodies, most notably the OECD.

2.ഩ Across the scenarios presented in this Outlook, the various policies that are assumed to be introduced and the different energy price levels that prevail could be expected to lead to some variations in GDP, as a result of the potentially important interactions of these variables on the economy. However, due to the uncertainty associated with estimating these effects and in order to more precisely identify the implications of different policy options on energy trends, the same level of GDP growth is assumed in each scenario. 38

World Energy Outlook 2013 | Global Energy Trends

Box 1.2 ‫ ٲ‬Uncertainties around the economic outlook

1

Risks to the economic recovery have eased in the last year, but momentum is proving to be slow to build. Global demand for products and services remains depressed, with sustained economic growth yet to take hold in many developed countries and growth forecasts being revised downward for many developing countries. In Japan, there has been a noƟceable recovery, although it remains to be seen how tapering of government sƟmulus spending and scheduled increases in sales tax will impact growth rates going forward. The Eurozone – as a whole – reported a return to modest growth in the second quarter of 2013, however many of its economies are sƟll experiencing low demand, relaƟvely high unemployment, weakness in the Įnancial sector and the eīects of austerity measures. Some countries remained in recession at the Ɵme of wriƟng, depressing energy demand: demand for both gas and electricity in Europe are at levels last seen in the early 2000s, with implicaƟons for energy sector revenues and investments. The economic crisis is also constraining the ability of governments to support a transiƟon to a low-carbon economy. The United States appears to be on a more promising growth trajectory, even having managed to weather the Įscal shock of automaƟc budget cuts relaƟvely well. However, the potenƟal tapering of central bank liquidity measures has caused some Įnancial market volaƟlity, which has been transmiƩed quickly through global bond and equity markets.

© OECD/IEA, 2013

Emerging economies have been the main engines of global growth over the past decade, but many are now facing slower growth at home on top of challenging economic condiƟons globally. China, in parƟcular, has played a deĮning role, contribuƟng more than three Ɵmes as much (35й) to global GDP growth as the United States (11й) over the years 2011-2012. But sustained high growth has also seen a build up of risks within China’s economy, including reliance on export-led growth, weaknesses in the Įnancial sector (both within and outside the formal banking sector) and local government Įnances, and concerns about the aīordability of property. China has policy levers available to support growth while also tackling vulnerabiliƟes in the system, i.e. to achieve a ͞soŌ landing͟, but achieving this is not a foregone conclusion. While China and the United States are the world’s largest energy consumers, China’s energy demand is much more sensiƟve to GDP trends. Given this and its growing oil, gas and coal imports, China’s economic outlook has potenƟally large repercussions for global energy markets. The assumed rates of growth in non-OECD countries imply that their combined GDP will surpass that of the OECD countries by around 2015͖ by 2035 their combined GDP will be 1.6ථƟmes larger. Some of the most rapid rates of growth are in developing Asia, which collecƟvely accounts for over half of the increase in global economic acƟvity during the period. China’s growth rate averages 5.7й in 2011-2035, despite falling aŌer 2020 to less than half the rate seen over the last decade, as its economy matures and its populaƟon

Chapter 1 | Scope and methodology

39

2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18

growth levels oī. Of all of the countries or regions that we have modelled as disƟnct enƟƟes, India’s GDP grows at the fastest rate, averaging 6.3й over the period. Brazil, the country focus in this Outlook (Part B), grows at 3.7й per year on average, well above our assumed rates of growth for the rest of LaƟn America. Table 1.2 ‫ ٲ‬Real GDP growth assumptions by region Compound average annual growth rate 1990-2011

2011-2015

2011-2020

2011-2035

OECD

2.2%

1.9%

2.2%

2.1%

Americas

2.5й

2.7й

2.9й

2.5й

2.4й

2.6й

2.8й

2.4й

Europe

2.0й

0.9й

1.5й

1.7й

Asia Oceania

1.9й

2.1й

2.1й

1.8й

0.9й

1.5й

1.4й

1.2й

Non-OECD

5.0%

5.6%

5.8%

4.8%

E. EuropeͬEurasia

0.7й

3.3й

3.5й

3.3й

0.6й

3.7й

3.6й

3.4й

United States

Japan

Russia Asia

7.5й

6.8й

7.1й

5.6й

China

10.0й

8.0й

8.1й

5.7й

India

6.5й

5.7й

6.5й

6.3й

5.0й

5.5й

5.5й

4.6й

Middle East

ASEAN

4.6й

3.2й

3.7й

3.7й

Africa

3.8й

5.1й

5.0й

4.0й

LaƟn America

3.4й

3.4й

3.7й

3.3й

3.0й

3.0й

3.6й

3.7й

World

3.3%

3.6%

4.0%

3.6%

European Union

1.8й

0.7й

1.3й

1.6й

Brazil

Note: Calculated based on GDP expressed in year-2012 dollars in purchasing power parity terms. Sources: IMF (2013)͖ OECD (2013)͖ Economist Intelligence Unit and World Bank databases͖ IEA databases and analysis.

© OECD/IEA, 2013

WŽƉƵůĂƟŽŶĂŶĚĚĞŵŽŐƌĂƉŚŝĐƐ PopulaƟon is a fundamental driver of energy demand, although the relaƟonship is not linear, as it depends on many other factors. Based on the medium variant of the latest United NaƟons projecƟons, the world populaƟon is set to rise from 7.0ථbillion in 2011 to 8.7ථbillion in 2035 (UNPD, 2013) (Tableථ1.3). Africa, India and Southeast Asia are the biggest contributors to the increase. By contrast, the populaƟon of China changes very liƩle over the period͖ by around 2025, India becomes the world’s most populous country. PopulaƟon growth is slow in OECD countries, although some see relaƟvely fast increases, including Australia, Canada, Chile, Mexico and the United States. The populaƟon of the United States increases by almost one-ĮŌh, underpinned by relaƟvely high levels of immigraƟon. As has been the case since the late 1960s, world populaƟon growth slows, falling from 1.2й in 2012 to 0.7й in 2035. 40

World Energy Outlook 2013 | Global Energy Trends

S P O T L I G H T

1

How does the IEA model future energy trends?

2

The IEA has used its World Energy Model (WEM) as the principal tool to generate the projecƟons that underpin the WEO scenarios for some two decades. The WEM is a large-scale simulaƟon model designed to replicate how energy markets funcƟon. It consists of three main modules: (i) Įnal energy consumpƟon͖ (ii) energy transformaƟon͖ and (iii) oil, natural gas, coal and bioenergy supply. Detailed, Ɵmely and reliable staƟsƟcs form a crucial input to the WEM. These are sourced primarily from the IEA’s historical staƟsƟcs on energy supply, trade, stocks, transformaƟon and demand, but are supplemented by addiƟonal data from governments, internaƟonal organisaƟons, energy companies, consulƟng Įrms and investment banks worldwide. Another crucial input is informaƟon on government policies that aīect energy demand and supply. The WEM is updated on an annual basis with new and more detailed features to ensure that it conƟnues to reŇect the changing dynamics of global energy markets and to enable greater disaggregaƟon of results.
sector, as well as to present internaƟonal trade Ňows of crude and oil products. „Coverage of the chemical and petrochemical sector has been improved to enable

energy consumpƟon and feedstock use to be modelled for each major product. „The modelling of eĸciency measures in the buildings sector (such as insulaƟon and

3 4 5 6 7 8 9 10 11 12

retroĮt programmes) has been revised and improved. „In preparaƟon of the WEO-2013 Special Report Southeast Asia Energy Outlook,

separate models have been built for Thailand, Malaysia and the Philippines (previously Indonesia was the only country in the region modelled on an individual basis) (IEA, 2013a). „ReŇecƟng the accession of CroaƟa to the European Union in July 2013, the models

© OECD/IEA, 2013

for Europe have been expanded to include all 28 member states. In June 2013, the IEA hosted the InternaƟonal Energy Workshop to bring together the leading analysts from all over the world to discuss the latest developments in energy and climate modelling. The agenda included a special session dedicated to the IEA’s WEM, which generated feedback and suggesƟons that are expected to enrich the IEA’s energy analysis in future years. More details on the WEM are available at www.worldenergyoutlook.org/weomodel/.

Chapter 1 | Scope and methodology

41

13 14 15 16 17 18

Urban areas are set to accommodate all of the growth in populaƟon, as the share of the world͚s populaƟon living in towns rises from 52й to 62й over 2011-2035. The number of people living in rural areas declines over the period. These changes will have implicaƟons for the amount and type of energy used. The concentraƟon of acƟviƟes in urban areas can facilitate improved energy eĸciency through economies of scale, however, people living in ciƟes and towns in developing countries typically use more energy than their rural counterparts. Other demographic changes taking place are also set to inŇuence energy demand paƩerns, most notably the rising share of older people and a decline in household size. These changes highlight the importance of long-term strategic planning to ensure that ciƟes and metropolitan areas develop in an energy-eĸcient manner. Table 1.3 ‫ ٲ‬Population assumptions by region PopulaƟon growthΎ 1990-2011 2011-2020 2011-2035 OECD

PopulaƟon (million) 2011 2035

UrbanisaƟon 2011

2035

0.7%

0.5%

0.4%

1 245

1 379

80%

86% 88й

1.1й

0.9й

0.8й

477

572

82й

1.0й

0.8й

0.7й

316

374

83й

88й

Europe

0.5й

0.4й

0.3й

563

600

75й

81й

Asia Oceania

0.4й

0.2й

0.0й

205

207

89й

93й

0.2й

-0.2й

-0.3й

128

118

91й

97й

Non-OECD

1.5%

1.2%

1.0%

5 715

7 322

46%

58%

E. EuropeͬEurasia

-0.1й

0.0й

-0.1й

337

327

63й

68й

-0.2й

-0.3й

-0.4й

142

129

74й

80й

1.3й

0.9й

0.7й

3 664

4 343

41й

55й

Americas United States

Japan

Russia Asia China

0.8й

0.5й

0.2й

1 351

1 431

51й

73й

India

1.7й

1.1й

0.9й

1 241

1 551

31й

42й 59й

1.4й

1.1й

0.9й

597

737

45й

Middle East

ASEAN

2.4й

1.9й

1.5й

209

297

67й

73й

Africa

2.4й

2.4й

2.3й

1 045

1 790

40й

51й

LaƟn America

1.4й

1.0й

0.8й

460

564

79й

84й

1.3й

0.8й

0.6й

197

226

85й

89й

World

1.3%

1.1%

0.9%

6 960

8 701

52%

62%

European Union

0.3й

0.2й

0.1й

508

518

74й

79й

Brazil

© OECD/IEA, 2013

Ύ The assumed compound average annual growth rates are the same for all scenarios presented in this Outlook. Sources: UNPD and World Bank databases͖ IEA analysis.

Growth in energy demand is closely correlated with growth in per-capita income, although the relaƟonship has decoupled in several advanced countries and may be weaker in the future in economies that are emerging today, should they ͞leapfrog͟ to more eĸcient energy-use pracƟces. Nonetheless, rising incomes will conƟnue to lead to increased demand for goods that require energy to produce and to use, such as cars, refrigerators and air condiƟoners. Vehicle ownership rates, for example, have historically taken oī once per-capita incomes pass a threshold of around Ψ4ථ000 to Ψ5ථ000. Based on our assumpƟons

42

World Energy Outlook 2013 | Global Energy Trends

for populaƟon and GDP growth, global GDP per capita is set to increase at 2.3й per year, from around Ψ10ථ300 in 2011 to around Ψ17ථ200 in 2035 (calculated using market exchange rate). GDP per capita will grow quickest in the developing countries, notably China and India, though in OECD countries it will sƟll be over Įve Ɵmes higher than the average of the rest of the world in 2035. China’s GDP per capita increases by three-and-a-half Ɵmes, surpassing the average of the Middle East around 2030 (Figureථ1.2). Figure 1.2 ‫ ٲ‬GDP per capita by region United States

1.7%

Japan

1.6%

European Union

4 5

3.0%

Middle East

7

2.4%

China

5.5%

ASEAN

8

3.7% 1.7% 5.3% 10

20

30

40

50

9

60 70 80 Thousand dollars (2012)

Notes: Calculated on the basis of GDP expressed in year-2012 dollars at market exchange rate. The percentages to the right of each bar represent the respecƟve compounded annual growth rates for the period 2011-2035.

Energy prices Prices aīect energy demand and supply through a wide variety of channels. The evoluƟon of energy pricing is, accordingly, a crucial determinant of future energy trends. On the demand side, it will aīect the amount of each fuel end-users choose to consume and their choice of technology and equipment to provide a parƟcular energy service. On the supply side, it will aīect producƟon and investment decisions.

© OECD/IEA, 2013

3

6

3.8%

Brazil

India

2035

2

1.5%

Russia

Africa

2011

1

The internaƟonal fossil fuel prices in each of the scenarios reŇect analysis of the price levels that would be needed to sƟmulate suĸcient investment in supply to meet projected demand over the period (Box 1.3). Average retail prices in end-uses, power generaƟon and other transformaƟon sectors in each region are derived from iteraƟve runs of the World Energy Model. These end-use prices take into account local market condiƟons, including taxes, excise duƟes, carbon-dioxide (CO2) emissions penalƟes and pricing, as well as any subsidies. In the three scenarios, the rates of value-added taxes and excise duƟes on fuels are assumed to remain unchanged, except where future tax changes have already been adopted or are planned. In the 450ථScenario, administraƟve arrangements (price controls or higher taxes) are assumed to be put in place to keep end-user prices for oil-based transport fuels at a level similar to those in the New Policies Scenario. Chapter 1 | Scope and methodology

43

10 11 12 13 14 15 16 17 18

Box 1.3 ‫ ٲ‬Deriving the fossil fuel prices used in WEO analysis The internaƟonal fossil fuel prices used in this report reŇect our judgement of the price levels that would be needed to sƟmulate suĸcient investment in supply to meet projected demand over the period. The resulƟng price trajectories are decepƟvely smooth: in reality prices are likely to be more volaƟle and cyclic. The price trajectories have been derived through an iteraƟve modelling exercise. First, the demand modules of the IEA’s World Energy Model (WEM) are run under a given set of prices (based on end-user prices). Once the resultant demand level is determined, the supply modules of the WEM calculate the levels of producƟon of oil, natural gas and coal that result from the given price levels, taking account of the costs of various supply opƟons and the constraints on producƟon rates of various types of resources (see Chapter 13 for a more detailed discussion in relaƟon to oil). In the event that the price is not suĸcient to generate enough supply to cover global demand, price levels are increased and a new level of demand and supply is quanƟĮed. This procedure is carried out repeatedly with prices adjusƟng unƟl demand and supply are in balance as a trend through the projecƟon period. In the near to medium term, the supply trajectories take into account our assessment of speciĮc individual projects that are currently operaƟng or have already been sancƟoned, planned or announced. For the longer term, they are consistent with our top-down assessment of the costs of exploraƟon and development of the world’s oil, natural gas and coal resources and our judgements of the feasibility and the rate of investment required in diīerent regions to turn these resources into producƟon.

© OECD/IEA, 2013

The price paths vary across the three scenarios presented in WEO-2013. In the Current Policies Scenario, policies adopted to reduce the use of fossil fuels are limited. This leads to higher demand and, consequently, higher prices, although prices are not high enough to trigger widespread subsƟtuƟon of fossil fuels by renewable energy sources. Lower energy demand in the 450ථScenario means that limitaƟons on the producƟon of various types of resources are less signiĮcant and there is less need to produce fossil fuels from resources higher up the supply cost curve. As a result, internaƟonal fossil fuel prices are lower than in the other two scenarios. However, this does not translate into lower end-user prices for oil-based transport fuels as price controls or higher taxes are assumed to keep them at a level similar to the New Policies Scenario. In the New Policies Scenario, subsidies to fossil fuel consumpƟon are phased out in all netenergy imporƟng countries within ten years at the latest. However, in net-energy exporƟng countries, they are phased out only if speciĮc policies to do so have been announced, in recogniƟon of the added diĸculƟes these countries are likely to face in reforming energy pricing. A survey undertaken for this report has idenƟĮed some 40 economies around the world that provide fossil-fuel consumpƟon subsidies. Within the group, the average rate of subsidisaƟon was 23й in 2012, meaning that consumers in those countries paid on average 77й of internaƟonal reference prices (see Chapterථ2). 44

World Energy Outlook 2013 | Global Energy Trends

Oil prices The world is experiencing a period of historically high oil prices. Brent crude oil has averaged over Ψ110ͬbarrel in real terms since 2011, a sustained period of high oil prices that is without parallel in oil market history. This has generated responses on the demand and supply sides. Higher oil prices have given consumers and industry extra incenƟve to improve energy eĸciency and have increased interest in subsƟtuƟng away from oil, for example to natural gas in road transport. Oil demand in the OECD is in decline. In the emerging economies, which have driven global demand, growth rates have slowed. The price rise has also led to increased interest in developing resources that were previously considered too diĸcult or too costly to produce. This is best exempliĮed by the spectacular rise in light Ɵght oil producƟon in the United States, and by growing interest in oil exploraƟon and producƟon in deepwater. Demand and supply side trends suggest that the global oil balance could ease over the next few years, despite concerns about oil supply security stemming from geopoliƟcal instability in parts of the Middle East and North Africa. In the 2020s, however, the balance is likely to shiŌ again, as non-OPEC supply levels oī and starts to decline.

2 3 4 5 6 7 8

In this Outlook, oil prices vary across the scenarios in line with the degree of policy eīort made to curb demand growth. In the New Policies Scenario, the average IEA crude oil import price – a proxy for internaƟonal oil prices – reaches Ψ113ͬbarrel (in year-2012 dollars) in 2020 and Ψ128ͬbarrel in 2035͖ the oil price picks up more quickly in the laƩer half of the period in line with Ɵghter market condiƟons (Tableථ1.4). In the Current Policies Scenario, substanƟally higher prices are needed to balance supply with faster growth in demand, reaching Ψ145ͬbarrel in 2035. In the 450 Scenario, lower oil demand means there is less need to produce oil from costly Įelds in non-OPEC countries, which are higher up the supply curve. As a result, the oil price peaks at around Ψ110ͬbarrel by 2020 and then falls slowly, reaching Ψ100ͬbarrel in 2035.

10

In Chapterථ14, the possibility of a Low Oil Price Case is examined, premised on supply developments in several countries turning out more posiƟvely than projected in the New Policies Scenario. In this case, output growth is rapid enough to ease the market balance, bringing on and meeƟng addiƟonal oil consumpƟon at a price that stabilises at Ψ80ͬbarrel.

13

Natural gas prices

© OECD/IEA, 2013

1

Although internaƟonal trade in natural gas conƟnues to expand rapidly, there is no global pricing benchmark for natural gas, as there is for oil. Rather, there are three major regional markets – North America, Asia-PaciĮc and Europe – with prices established by diīerent mechanisms. In North America, gas trade relies on hub-based pricing, with prices reŇecƟng local gas supply and demand. In Asia-PaciĮc, trade is dominated by long-term contracts in which prices are at least partly indexed to the price of oil. Gas trade in Europe is gradually moving to gas-to-gas compeƟƟon, though about half of European trade today is governed by long-term oil-indexed contracts. A notable excepƟon in Europe is the United
45

9

11 12

14 15 16 17 18

There have always been diīerences in natural gas prices across the three major markets, reŇecƟng primarily their diīerent demand and supply balances and pricing systems. Since mid-2008, the gap has widened considerably. Prices in North America have fallen thanks to spectacular growth in shale gas output and reduced demand (owing to the economic crisis). By contrast, prices in Asia-PaciĮc (and to a lesser extent Europe) have risen, mostly due to the prevalence of oil-price indexaƟon at a Ɵme of persistently high oil prices. In 2012, average natural gas prices in the United States were less than one-quarter of the prices in Europe and one-sixth of those in Japan. By mid-2013, however, spot prices for gas at Henry Hub – the leading trading hub in the United States – had more than doubled from the lows reached in early 2012, narrowing regional price divergences. Nonetheless, low prices conƟnue to generate strong interest in exporƟng liqueĮed natural gas (LNG) from North America and raise quesƟons about the long-term sustainability of oil-linked pricing mechanisms.

Dollars per MBtu (2012)

Figure 1.3 ‫ ٲ‬Natural gas prices by region in the New Policies Scenario 18 Japan

15

Europe

12

2.2x 6.2x

9

United States 6 3

© OECD/IEA, 2013

1990

2000

2010

2020

2030

2035

In WEO-2013, large geographical spreads in natural gas prices persist during the Outlook period, albeit with a degree of convergence brought about by rising LNG supplies, increased short-term trading and greater operaƟonal Ňexibility (Figureථ1.3). These developments allow price changes in one part of the world to be reŇected more quickly elsewhere, but are unlikely to be suĸcient to create a single global price for gas, parƟcularly given the signiĮcant costs associated with liquefacƟon and shipping. In each of the scenarios, North American prices are lowest, reŇecƟng abundant and relaƟvely low-cost unconvenƟonal resources. But prices rise in absolute terms and relaƟve to the other regions, parƟcularly later in the period, as the costs of unconvenƟonal gas producƟon increases and as oil indexaƟon loosens gradually in other markets, notably Europe, as long-term contracts expire and are renegoƟated. In the New Policies Scenario, gas prices in 2035 are Ψ6.8ථper million BriƟsh thermal units (MBtu) (in year-2012ථdollars) in North America, Ψ12.7ͬMBtu in Europe and Ψ14.9ͬMBtu in Asia-PaciĮc. Prices in Japan are more than double those in the 46

World Energy Outlook 2013 | Global Energy Trends

United States in 2035, meaning that the spread is much narrower than observed recently, but much greater than before US producƟon of shale gas took oī. Gas prices vary across the three scenarios in line with the degree of policy eīort to curb CO2 emissions. The Gas Price Convergence Case, presented in Chapter 3, invesƟgates the condiƟons under which convergence between pricing mechanisms and prices could be more pronounced than in the New Policies Scenario. The case rests on three main condiƟons: (i) a larger volume of LNG export from North America͖ (ii) new supply contracts weakening or breaking the link with oil-price indexaƟon, and an accelerated pace of regulatory change for the gas sector across the Asia-PaciĮc region͖ and (iii) an easing of costs of construcƟng liquefacƟon plants and of shipping LNG. Compared with the New Policies Scenario, gas prices are slightly higher in North America but lower in Europe and in the Asia-PaciĮc region. The diīerenƟal between the US price and the European import price narrows to Ψ4.5ͬMBtu, with an extra Ψ1ͬMBtu to Asia-PaciĮc reŇecƟng addiƟonal transport costs.

The global coal market consists of various regional sub-markets that are typically separated by geography, coal quality or infrastructure constraints. As a result, coal prices vary markedly across the regions. InternaƟonal coal trade is a comparaƟvely small sub-market, yet it links various domesƟc markets through imports, exports and price movements. The degree to which regional coal prices Ňuctuate with price movements on the internaƟonal market depends on how well they are connected to it. Around one-ĮŌh of global steam coal producƟon is traded internaƟonally, with the remainder used closer to where it is mined. InternaƟonal trade has historically been divided into two market areas – Asia-PaciĮc and AtlanƟc – reŇecƟng the wide geographical spread of producƟon and the signiĮcance of transportaƟon costs as a share of the total delivered cost of coal. However, trade between the two market areas is growing, owing to increased supply sources and lower freight costs. The market in internaƟonally traded coal is dominated by spot market transacƟons, though long-term contracts with prices Įxed annually remain important in some cases.

© OECD/IEA, 2013

2 3 4 5 6 7

Steam coal prices

Prices of internaƟonally traded coal have Ňuctuated widely over the last decade. Strong demand saw prices climb throughout the early 2000s to record highs, above Ψ200ͬtonne in mid-2008. Prices then plummeted in the wake of the global economic crisis, before staging a recovery, underpinned by robust demand and weather-related supply constraints in a number of key producer countries. Since mid-2011, prices have again fallen, on weak demand and growing supply in the market and, by mid-2013, they were less than half their peak of 2008. Coal consumpƟon in China has been subdued because of slower growth in its electricity demand and increased hydropower output. As the world’s largest coal buyer, China exerts a major inŇuence on internaƟonal prices. The boom in US unconvenƟonal gas producƟon has also been a factor in depressing internaƟonal coal prices: some of the coal displaced by cheaper gas in US power generaƟon has found its way onto export markets.

Chapter 1 | Scope and methodology

1

47

8 9 10 11 12 13 14 15 16 17 18

© OECD/IEA, 2013

48

Table 1.4 ‫ ٲ‬Fossil fuel import prices by scenario (dollars per unit) New Policies Scenario

Current Policies Scenario

450 Scenario

Unit

2012

2020

2025

2030

2035

2020

2025

2030

2035

2020

2025

2030

2035

barrel

109

113

116

121

128

120

127

136

145

110

107

104

100

















Real terms (2012 prices) IEA crude oil imports Natural gasථ

World Energy Outlook 2013 | Global Energy Trends

United States

MBtu

2.7

5.1

5.6

6.0

6.8

5.2

5.8

6.2

6.9

4.8

5.4

5.7

5.9

Europe imports

MBtu

11.7

11.9

12.0

12.3

12.7

12.4

12.9

13.4

14.0

11.5

11.0

10.2

9.5

Japan imports

MBtu

16.9

14.2

14.2

14.4

14.9

14.7

15.2

15.9

16.7

13.4

12.8

12.2

11.7

tonne

99

106

109

110

110

112

116

118

120

101

95

86

75

















136

156

183

216

132

144

157

169

















OECD steam coal imports Nominal terms IEA crude oil imports

barrel

109

Natural gas

144

171

205

245

United States

MBtu

2.7

6.1

7.5

9.1

11.6

6.2

7.7

9.3

11.7

5.8

7.2

8.6

10.0

Europe imports

MBtu

11.7

14.2

16.1

18.5

21.5

14.9

17.3

20.2

23.6

13.8

14.7

15.4

16.0

Japan imports

MBtu

16.9

17.1

19.1

21.7

25.1

17.7

20.4

24.0

28.2

16.1

17.2

18.4

19.7

tonne

99

127

146

165

186

134

155

178

202

121

128

129

127

OECD steam coal imports

Notes: Gas prices are weighted averages expressed on a gross caloriĮc-value basis. All prices are for bulk supplies exclusive of tax. The US price reŇects the wholesale price prevailing on the domesƟc market. Nominal prices assume inŇaƟon of 2.3й per year from 2012.

The outlook for coal prices depends heavily on the stringency of climate policy measures and competition between natural gas and coal in power generation. International steam coal prices (which are used to derive prices for coking coal and other coal qualities) vary markedly across the three scenarios. In the New Policies Scenario, the average OECD steam coal import price reaches Ψ106ͬtonne (in year-2012 dollars) in 2020, from its average of Ψ99ͬtonne in 2012, before rising slowly to about Ψ110ͬtonne in 2035. Coal prices rise more quickly in the Current Policies Scenario, on stronger demand growth, but fall sharply in the 450 Scenario, reflecting the impact of much stronger policy action to reduce CO2 emissions.

The last year has seen an increase in the number of schemes that put a price on carbon emissions. The EU Emissions Trading System (ETS) remains the world’s largest scheme, covering all 28 member states of the European Union, plus Norway, Iceland and Liechtenstein. Programmes are also in place in New ealand, Australia, California (United States), Quebec and Alberta (Canada) and
© OECD/IEA, 2013

2 3 4 5

Carbon markets

Our assumpƟons on carbon pricing vary across the scenarios, reŇecƟng the diīerent levels of policy intervenƟon to curb growth in CO2 emissions. We assume each of the exisƟng and planned programmes that are described above conƟnue, with the price of CO2 rising under each programme over the projecƟon period (Tableථ1.5). In Europe, the price increases from an average of Ψ10ͬtonne (in year-2012 dollars) in 2012 to Ψ20ͬtonne in 2020 and Ψ40ͬtonne in 2035. A CO2 price covering all sectors is introduced in China starƟng in 2020, Chapter 1 | Scope and methodology

1

49

6 7 8 9 10 11 12 13 14 15 16 17 18

© OECD/IEA, 2013

50

Figure 1.4 ‫ ٲ‬Current and proposed schemes that put a price on carbon Canada

Quebec Start year 2013

European Economic Area

Switzerland Start year 2008

Start year 2005

Sectors

Sectors

Electricity and industry

Sectors

Electricity and industry Alberta Start year 2007 Sectors Sectors Sectors Sectors

Electricity, industry and aviation

Kazakhstan

Industry Ontario To be determined Manitoba To be determined British Columbia To be determined

Beijing, Tianjin, Shanghai Chongqing, Shenzhen

Provinces

Guangdong, Hubei

Sectors

Sectors

Electricity and industry Sectors

World Energy Outlook 2013 | Global Energy Trends

Tokyo Start year 2010

Korea

Electricity and industry Regional GHG Initiative Start year 2009

Vary by pilot scheme National trading system To be determined

Japan

Ukraine Sectors To be determined

Sectors

Sectors

Cities

Start year 2013

United States of America

California Start year 2013

China

Pilots in cities and provinces Start year 2013

Start year 2015 Sectors

Electricity and industry

Electricity

Sectors

Commercial buildings and industry Saitama Start year 2011

Sectors

Commercial buildings and industry

Mexico Sectors To be determined

Chile

Brazil

Rio de Janeiro Sectors To be determined

Sectors To be determined

In place Implementation scheduled

New Zealand

Australia

South Africa Start year 2015

Start year 2012

Start year 2008

Sectors

Sectors

Sectors

Industry with partial exemptions for certain sub-sectors

Electricity, industry, waste, forestry, domestic aviation and shipping

Under consideration This map is without prejudice to the status of or sovereignty over any territory, to the delimitation of international frontiers and boundaries, and to the name of any territory, city or area.

Electricity, industry, waste, forestry, transport fuels and domestic aviation

starƟng at Ψ10ͬtonne and then rising to Ψ30ͬtonne in 2035. As a result of these schemes, the share of global CO2 emissions subject to a carbon price increases from around 8й in 2012 to around one-third in 2035. This result is parƟcularly sensiƟve to our assumpƟon that a scheme is introduced in China, without which the share would drop to 6й in 2035. We also assume that, from 2015 onwards, all investment decisions in the power sector in Canada, the United States and Japan include an implicit or ͞shadow͟ price for carbon. In general, the CO2 price levels assumed in WEO-2013 are lower than in WEO-2012, reŇecƟng the low prices over the past year and lower expectaƟons in the longer term. Table 1.5 ‫ ٲ‬CO2 price assumptions in selected regions by scenario

Current Policies Scenario

New Policies Scenario

450 Scenario

Sectors

European Union Australia and New ealand

2020

2030

2035

Power, industry and aviaƟon

15

25

30

AllΎ

15

25

30


Power and industry

15

25

30

European Union

Power, industry and aviaƟon

20

33

40

Australia and New ealand

AllΎ

20

33

40


Power and industry

20

33

40

China

All

10

24

30

South Africa

Power and industry

8

15

20

United States and Canada

Power and industry

20

95

125

European Union

Power, industry and aviaƟon

35

95

125

Japan

Power and industry

20

95

125


Power and industry

35

95

125

Australia and New ealand

All

35

95

125

China, Russia, Brazil and South Africa

Power and industryΎΎ

10

70

100

Ύ Agriculture is not assumed to be included in New ealand’s Emissions Trading Scheme. ΎΎ All sectors in China. Note: In the New Policies Scenario, a shadow price for CO2 in the power sector is assumed to be adopted as of 2015 in the United States, Canada and Japan (starƟng at Ψ15ͬtonne and rising to Ψ35ͬtonne in 2035).

© OECD/IEA, 2013

Technology Successive ediƟons of the WEO have demonstrated the need for ongoing improvements in eĸciency, including energy conservaƟon and management, and the adopƟon of a porƞolio of exisƟng and new technologies in order to address the challenges posed by the world’s rising fossil energy use. It follows that the rate at which energy eĸciency improves and new technologies for supplying and using energy are developed and deployed will have a major

Chapter 1 | Scope and methodology

2 3 4 5

(in year-2012 dollars per tonne) Region

1

51

6 7 8 9 10 11 12 13 14 15 16 17 18

impact on future energy balances, both in terms of the overall amount of energy used and the fuel mix. An IEA review released in mid-2013 concluded that recent progress in developing and deploying clean energy technologies and in improving energy eĸciency has not been suĸcient to achieve announced policy objecƟves and is being limited by market failures (IEA, 2013c). But it saw some reasons for opƟmism. For example, annual sales of hybrid vehicles in 2012 passed the 1ථmillion mark for the Įrst Ɵme and solar photovoltaic (PV) systems and wind turbines were installed at a rapid pace by historical standards (Tableථ1.6). Table 1.6 ‫ ٲ‬Recent progress and key conditions for faster deployment of

© OECD/IEA, 2013

clean energy technologies Technology

Recent progress

Key condiƟons for faster deployment

Renewable power

Investment fell by 11й in 2012 from 2011 due to tougher Įnancing condiƟons, policy uncertainty and falling technology costs. Solar PV capacity sƟll grew by 42й and wind by 19й, compared with 2011 cumulaƟve levels.

Ongoing subsidies (as renewables generally remain more expensive than other sources of power). Reforms to facilitate grid integraƟon. Increased RDΘD in emerging technologies, such as concentraƟng solar power, ocean and enhanced geothermal.

Nuclear power

Seven projects started construcƟon in 2012, an increase from 2011 when new projects fell to only four aŌer the Fukushima Daiichi accident. In 2010 there were 16 new projects.

More favourable electricity market mechanisms and investment condiƟons to reduce risk and allow investors to recover high upfront capital costs. Quick implementaƟon of post-Fukushima safety upgrades to foster public conĮdence.

Carbon capture and storage (CCS)

13 large-scale CCS demonstraƟon projects are in operaƟon or under construcƟon. ConstrucƟon began on two new integrated projects in 2012, while eight projects were cancelled.

Financial and policy commitment by governments to accelerate demonstraƟon eīorts. Suĸciently high price on CO2 emissions or a commercial market for captured CO2 for enhanced oil recovery.

Biofuels

New investment was 50й lower in 2012 than in 2011, as a result of overcapacity, and a review of biofuels support policies and higher feedstock prices.

A longer-term policy framework to build investor conĮdence. RDΘD to improve cost and eĸciency, and to develop sustainable feedstocks. Development and applicaƟon of internaƟonally agreed sustainability criteria and standards.

Hybrid (HV) and HV sales reached 1.2 million in 2012, electric vehicles up 43й on 2011, led by Japan and (EV) the United States. EV sales more than doubled from 2011 to 2012, from a low base. Government targets for EV sales increased.

Further reducƟons in baƩery costs and other measures to enhance compeƟƟveness. Non-Įnancial incenƟves, such as priority access to parking and restricted highway lanes. InstallaƟon of recharging infrastructure.

Energy eĸciency

Policy acƟon to remove the barriers obstrucƟng the implementaƟon of energy eĸciency measures that are economically viable (see Chapter 7).

Evidence of renewed focus from governments, with many major energyconsuming countries announcing new measures.

Sources: IEA (2013c and 2013d).

52

World Energy Outlook 2013 | Global Energy Trends

WEO-2012 found that even though there is a renewed policy focus on energy eĸciency, two-thirds of the economic potenƟal to improve energy eĸciency is set to remain untapped in the period to 2035. While investment in many energy-eĸcient technologies and pracƟces appear to make good economic sense, the level of their deployment is oŌen much lower than expected due to the persistence of a number of barriers.
© OECD/IEA, 2013

AmbiƟous carbon abatement also necessitates a shiŌ to low-carbon fuels in the transport sector, as vehicle fuel-economy improvements alone will not lead to the steep emissions reducƟons required. While natural gas and biofuels are promising alternaƟves to oil, their potenƟal to reduce emissions, relaƟve to oil, is limited, owing to their carbon content (natural gas) or concerns about their sustainability and conŇicts over land use or other uses for the feedstock (convenƟonal biofuels). High expectaƟons rest on the deployment of electric and plug-in hybrid electric vehicles. But increasing their market penetraƟon will require major cost reducƟons and addressing issues crucial to consumer acceptability, such as driving range (for example, through fast-recharging infrastructure). In each of the scenarios presented in this Outlook, energy technologies – both on the demand and supply sides – that are in use today or are approaching commercialisaƟon are assumed to achieve ongoing cost reducƟons as wider deployment contributes to more eĸcient producƟon. No complete technological breakthroughs are assumed to be made, as it cannot be known what they might involve, whether or when they might occur and how quickly they might be commercialised. The pace of eĸciency gains for end-

Chapter 1 | Scope and methodology

53

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18

© OECD/IEA, 2013

use technologies varies for each fuel and each sector, depending on our assessment of the potenƟal for improvements and the stage reached in technology development and commercialisaƟon. Technological advances are also assumed to improve the eĸciency of producing and supplying energy. For many regions and technologies, energy derived from renewable sources is today more costly than energy from fossil fuels and therefore requires subsidies in order to aid its deployment (see Chapterථ6). We assume that exisƟng subsidies for renewable energy technologies are retained unƟl suĸcient cost reducƟons have been achieved to enable them to compete on their own merits with convenƟonal technologies. At that point, we assume subsidies cease to be awarded to addiƟonal producƟon.

54

World Energy Outlook 2013 | Global Energy Trends

Chapter 2 Global energy trends to 2035 Finding our way in a new energy world Highlights

x Global energy demand will grow to 2035, but government policies can inŇuence the pace. In the New Policies Scenario, our central scenario, global energy demand increases by one-third from 2011 to 2035. Demand grows for all forms of energy: oil by 13%, coal by 17% (mainly before 2020), natural gas by 48%, nuclear by 66% and renewables byථ77%. Energy-related CO2 emissions rise by 20%, reaching 37.2ථGt.

x Emerging economies account for more than 90% of global net energy demand growth, but this comes from mulƟple and someƟmes unexpected sources. While Asian energy demand growth is led by China this decade, it shiŌs towards India and, to a lesser extent, Southeast Asia aŌer 2025.ථThe Middle East emerges as a major energy consumer, with its gas demand growing by more than the enƟre OECD: the Middle East is the second-largest gas consumer by 2020 and third-largest oil consumer by 2030, redeĮning its role in global energy markets.

x Electricity demand grows by more than any other Įnal form of energy. Although its share declines, coal conƟnues to be the largest source of electricity generaƟon and coal-gas price dynamics remain important for regional trends. Nearly half of the net increase in electricity generaƟon comes from renewables and their share of the total reaches more than 30% by 2035. Diīerent natural gas and electricity prices across regions conƟnue to have implicaƟons for relaƟve industrial compeƟveness.

x World oil demand grows from 87 mbͬd in 2011 to 101 mbͬd in 2035, with transport and petrochemicals being key drivers. One-third of the net global growth fuels Asia’s road freight. The reĮning industry faces huge structural challenges: the composiƟon of feedstocks changes, while oil product demand shiŌs towards Asia and the Middle East, and towards diesel, naphtha and kerosene. Global reĮning capacity grows by 13ථmbͬd to 2035, but some regions risk being leŌ with substanƟal idle capacity.

x Non-OPEC supply plays the major role in meeƟng net oil demand growth this decade, but OPEC plays a far greater role aŌer 2020. The United States is the world’s largest oil producer from 2015 to the early 2030s͖ light Ɵght oil and eĸciency policies reduce rapidly its reliance on imports. Brazil becomes a major oil exporter, delivering one-third of global supply growth to 2035. China is about to become the largest oil importer and becomes the largest oil consumer around 2030. The European Union stays the largest gas importer, but demand returns to 2010 levels only as 2035 approaches.

© OECD/IEA, 2013

x Despite some signs of reform, fossil-fuel subsidies increased to Ψ544ථbillion in 2012. Subsidies to renewables increased by 11% to reach Ψ101ථbillion. Nearly 1.3ථbillion people did not have access to electricity in 2011 and more than 2.6ථbillion relied on the tradiƟonal use of biomass for cooking. More than 95% of these people are in Asia or sub-Saharan Africa, and they are mainly in rural areas. Chapter 2 | Global energy trends to 2035

55

Overview of energy trends by scenario Many of the long-held tenets of the energy sector are being rewriƩen. Major importers are becoming exporters, large exporters are becoming large consumers and previously small consumers are becoming the dominant source of global demand. These changes emerge as the energy sector acts and reacts to broader global trends, such as shiŌs in economic growth, demographic change, industrialisaƟon, electriĮcaƟon, eīorts at decarbonisaƟon, technological breakthroughs and divergent regional energy prices. The energy sector itself is innovaƟng at a rapid pace: unlocking unconvenƟonal oil and gas supplies, enhancing supply Ňexibility with liqueĮed natural gas (LNG), integraƟng larger shares of variable renewable supply into the power sector and increasing energy eĸciency. Our understanding of the energy sector must therefore evolve if we are to take the best policy and investment decisions. This ediƟon of the tŽƌůĚ ŶĞƌŐLJ KƵƚůŽŽŬ෴;tKͲϮϬϭϯͿ seeks to put the latest developments into perspecƟve and explore their implicaƟons for global energy security, economic development and the environment. tKͲϮϬϭϯ takes 2011 to 2035 as its Outlook periodථand considers three scenarios based on diīering policy assumpƟons (see Chapterථ1)͖ the results vary signiĮcantly (Boxථ2.1). The New Policies Scenario – our central scenario – takes account of exisƟng policies and the anƟcipated impact of the cauƟous implementaƟon of declared policy intenƟons. The Current Policies Scenario takes account only of policies enacted as of mid-2013, providing a baseline of how global energy markets would evolve if established trends conƟnue unabated. The 450ථScenario illustrates an energy pathway compaƟble with a 50% chance of limiƟng the long-term increase in average global temperature to 2ථdegrees Celsius (ΣC).

© OECD/IEA, 2013

Box 2.1 ‫ ٲ‬Building on a new base A reŇecƟon on the 2011 base year for tKͲϮϬϭϯ and how it has changed from 2010 is useful before examining the key Įndings from the projecƟons. Total primary energy demand increased by around 1.4% in 2011, compared with a robust 5.6% increase the year before (a year of economic rebound). Within this global trend, demand declined in Japan by 7.5%, in the European Union by around 3.5% and in the United States by just over 1% (although US coal demand was down nearly 5%). The exact drivers were country-speciĮc, but a weak global economy, the repercussions of the Fukushima Daiichi nuclear accident in Japan, high fuel costs (in some cases), eīorts to improve eĸciency and the weather were among them. In contrast, primary energy demand increased by 8% in China, nearly 4% in
World Energy Outlook 2013 | Global Energy Trends

A growing global populaƟon and expanding economy will conƟnue to push primary energy demand higher, but government policies will play an important role in dictaƟng the paceථ(Figureථ2.1). In the New Policies Scenario, global primary energy demand increases by one-third between 2011 and 2035, reaching around 17ථ400ථmillion tonnes of oil equivalentථ(Mtoe). Demand rises more quickly in the Current Policies Scenario, ending nearly 45% higher than 2011, equivalent to adding the combined energy demand of the world’s three largest consumers today (China, the United States and India). In both cases, energy demand grows most rapidly in this decade and moderates aŌer 2020. Energy demand grows much more slowly in the 450 Scenario, increasing by only 14% over the Outlook period, and just 0.3% per year aŌer 2020, which, given historical rates of global energy growth, would represent a massive and extremely challenging change in trajectory. The non-OECD share of global energy demand has increased from 45% in 2000 to 57% inථ2011. This trend conƟnues, reaching around 60% in 2020 and around two-thirds in 2035 in each scenario. Compared with tKͲϮϬϭϮ, global energy demand in 2035 is 0.2% lower in the Current Policies Scenario, 1.1% higher in the New Policies Scenario and 0.8% higher in the 450ථScenario.

1

Figure 2.1 ‫ ٲ‬World primary energy demand and related CO2 emissions by

8

2 3 4 5 6 7

20 000

80

15 000

60

Gt

Mtoe

scenario

9

Primary energy demand: Current Policies Scenario

10

New Policies Scenario 450 Scenario

10 000

11

40 CO2 emissions (right axis):

5 000

20

Current Policies Scenario New Policies Scenario

13

450 Scenario 1990

2000

2010

2020

2030 2035

14

© OECD/IEA, 2013

Note: Mtoe с Million tonnes of oil equivalent͖ Gt с gigatonnes.

Fossil fuels account for 82% of primary energy demand in 2011, but the share in 2035 declines in all scenarios: to 76% in the New Policies Scenario, 80% in the Current Policies Scenario and 64% in the 450ථScenario, showing that, even in a 2ථΣC climate scenario, the transiƟon away from fossil fuels is likely to take considerable Ɵme to achieveථ(Tableථ2.1). The future trends diīer markedly by fuel. Demand for natural gas grows in all scenarios and, in absolute terms, increases more than all other fuels in the New Policies Scenario. Its relaƟve abundance, Ňexibility as a fuel and lower emissions than other fossil fuels all contribute to its relaƟvely bright outlook. In contrast, the demand for coal swings from seeing the largest increase in demand (44%) in the Current Policies Scenario to the largest decrease (33%) in Chapter 2 | Global energy trends to 2035

12

57

15 16 17 18

the 450ථScenario, reŇecƟng the considerable range of uncertainty resulƟng from diīerent policy paths. In the Current Policies Scenario, coal overtakes oil in the early 2020s as the largest fuel in the energy mix, while in the 450ථScenario coal demand drops below that of natural gas in the mid-2020s. Oil also has mixed results across scenarios, inŇuencing the speed at which new supply will need to be brought onlineථ(see Part C for a detailed Outlook for oil markets). Table 2.1 ‫ ٲ‬World primary energy demand and energy-related CO2 emissions by scenario New Policies Scenario

Current Policies Scenario

450 Scenario

2000

2011

2020

2035

2020

2035

2020

2035

Coal

2 357

3 773

4 202

4 428

4 483

5 435

3 715

2 533

Oil

3 664

4 108

4 470

4 661

4 546

5 094

4 264

3 577

Gas

2 073

2 787

3 273

4 119

3 335

4 369

3 148

3 357

Nuclear

676

674

886

1 119

866

1 020

924

1 521

Hydro

225

300

392

501

379

471

401

550

1 016

1 300

1 493

1 847

1 472

1 729

1 522

2 205

60

127

309

711

278

528

342

1 164

10 071

13 070

15 025

17 387

15 359

18 646

14 316

14 908

Bioenergy* Other renewables Total (Mtoe) Fossil fuel share

ϴϬй

ϴϮй

ϴϬй

ϳϲй

ϴϬй

ϴϬй

ϳϴй

ϲϰй

EŽŶͲKƐŚĂƌĞΎΎ

ϰϱй

ϱϳй

ϲϭй

ϲϲй

ϲϭй

ϲϲй

ϲϬй

ϲϰй

CO2 emissions (Gt)

23.7

31.2

34.6

37.2

36.1

43.1

31.7

21.6

© OECD/IEA, 2013

*ථIncludes tradiƟonal and modern biomass uses. **ථExcludes internaƟonal bunkers.

While consƟtuƟng a relaƟvely small share of the energy mix today (13% in 2011), global demand for renewable energy increases strongly to 2035 in all scenarios, by around 75% in the New Policies Scenario, nearly 60% in the Current Policies Scenario and more than 125% in the 450ථScenario. Policies already implemented, including subsidies, have given a boost to renewables and those adopted but yet to be implemented give a further push in the New Policies Scenario͖ but addiƟonal policies, oŌen targeted at objecƟves such as energy security or tackling environmental concerns, would see the penetraƟon of renewables increase substanƟally in the 450ථScenario. The outlook for hydropower varies liƩle across the scenarios, reŇecƟng the extent to which it is driven by the intenƟons and technically exploitable resources of a small number of countries, such as China and Brazilථ(Figureථ2.2). The main diīerence between scenarios occurs in the uptake of bioenergy and other renewables, such as wind and solar which, while cost compeƟƟve in some countries, require conƟnued government support in a number of cases in order to sƟmulate increased adopƟon. Taking into account nuclear power, which increases in all scenarios, low-carbon energy meets less than one-quarter of the growth in primary energy demand in the Current Policies Scenario, around 40% of the growth in the New Policies Scenario and more than 80% of the increase (of those energy sources whose demand rises) in the 450ථScenario.

58

World Energy Outlook 2013 | Global Energy Trends

Mtoe

Figure 2.2 ‫ ٲ‬Change in world primary energy demand by scenario, 2011-2035 2 000

CPS

1 500

NPS 450

1 000

1 2 3

500

4

0 -500

5

-1 000 -1 500 Coal

Oil

Gas

Nuclear

Hydro

Bioenergy

6

Other renewables

7

Note: CPS с Current Policies Scenario͖ NPS с New Policies Scenario͖ 450 с 450ථScenario.

There is a growing disconnect between the greenhouse-gas emissions trajectory that the world is on and one that is consistent with the 2ථΣC climate goal. The energy sector accounts for more than two-thirds of global greenhouse-gas emissions (IEA,ථ2013a) and, in 2012, we esƟmate that energy-related carbon dioxide (CO2) emissions increased by 1.2% to 31.5ථgigatonnes (Gt). The scenarios have a signiĮcantly diīerent impact on the level of future emissions. By 2035, global energy-related CO2 emissions are projected to increase to 37.2ථGt in the New Policies Scenario and 43.1ථGt in the Current Policies Scenario, but they decrease to 21.6ථGt in the 450 Scenario.2 In the absence of addiƟonal policies, as in the Current Policies Scenario, CO2 emissions would be twice the level in the 450ථScenario in 2035, while the cauƟous implementaƟon of announced policies, as in the New Policies Scenario, achieves nearly 30% of the cumulaƟve savings needed to be on a trajectory consistent with limiƟng the average global temperature rise to 2ථΣC.

Energy demand

10 11 12

14

In the New Policies Scenario, global energy demand grows by 1.6% per year on average to 2020 and then gradually slows to average 1% per year thereaŌer, reaching around 17ථ400ථMtoe in 2035ථ(Figureථ2.3). Associated with this 33% increase in energy demand over the projecƟon period, the global populaƟon grows by around one-quarter and the global economy more than doubles. Energy demand growth slows primarily as a result of a gradual slowdown in economic growth in certain countries, parƟcularly the largest rapidly industrialising developing economies, and as recently announced energy policies © OECD/IEA, 2013

9

13

Energy trends in the New Policies Scenario

2.ഩ See the WEO special report ZĞĚƌĂǁŝŶŐƚŚĞŶĞƌŐLJͲůŝŵĂƚĞDĂƉ (IEA, 2013a) and the Spotlight in this chapter for more on the pragmatic and economic actions the energy sector can take to keep open the path to a 2ථΣC climate trajectory.

Chapter 2 | Global energy trends to 2035

8

59

15 16 17 18

(targeted at increasing energy security, improving eĸciency and reducing polluƟon) are implemented and have a greater eīect over Ɵme. Despite these acƟons, global energy demand is 190ථMtoe higher in 2035 than projected last year. In the OECD, a comparison with tKͲϮϬϭϮ shows demand in 2035 to be slightly lower across all fuels, mainly as a result of the conƟnuing economic woes in many countries. In contrast, non-OECD energy demand is generally higher, the biggest change being higher coal demand in 2035, mainly due to an upward revision of coal used as petrochemical feedstock in China (see Chapterථ15). Figure 2.3 ‫ ٲ‬Primary energy demand and energy intensity in the New 0.5

16 000

0.4

Mtoe

20 000

12 000

0.3

8 000

0.2

4 000

0.1

1980

1990

2000

2010

2020

toe per thousand dollars of GDP ($2012, MER)

Policies Scenario Energy demand: Non-OECD OECD Energy intensity (right axis): Non-OECD OECD

2030 2035

Note: toe с tonne of oil equivalent͖ MER с market exchange rate.

© OECD/IEA, 2013

A renewed focus on energy eĸciency, at a Ɵme of relaƟvely high energy prices, has accelerated the previously slow rate of improvement in global energy intensityථ(see Chapter 7).3 From 2000 to 2010, the amount of energy used to produce a unit of gross domesƟc product (GDP) declined by 0.4% per year on average. But there has been a signiĮcant improvement since 2010 and, in 2012, the amount of energy used to produce a unit of GDP declined by 1.5%. This has been driven by high energy prices inducing energy conservaƟon, renewed government-led acƟon in support of energy eĸciency and fuel switching. The long-term improvement in global energy intensity is expected to conƟnue through the projecƟon period – energy intensity is down by more than one-third in 2035. Energy eĸciency policies, a primary contributor to energy intensity improvements in the New Policies Scenario, deliver global savings of 910ථMtoe in 2035, compared with the Current Policies Scenario, a level equivalent to slightly more than half the current energy use of the European Union. In cumulaƟve terms, these eĸciency-related primary energy savings are more than 9ථ200ථMtoe over the projecƟon period. China sees the biggest eĸciency gains in the New Policies Scenario (relaƟve to the Current Policies Scenario), 3.ഩ Energy intensity is often used as a proxy measure – albeit an imperfect one – for energy efficiency. It is calculated as primary energy demand per dollar of GDP at market exchange rate. 60

World Energy Outlook 2013 | Global Energy Trends

as policies, such as those in its 12thථFive-zear Plan, deliver important improvements. The United States also makes signiĮcant gains as a result of its energy eĸciency policies. In 2035, industry accounts for 37% of total eĸciency-related energy savings globally and buildingsථfor 26%. In both sectors, the bulk of the savings are made in the use of electricity, led by eĸciency improvements in electric motor systems, stricter standards for appliances and more eĸcient lighƟng. In the transport sector, improved fuel-economy standards lead to oil savings of around 5ථmillion barrels per day (mbͬd) by 2035. Improvements in the eĸciency of fossilථfuel-Įred power plants account for most of the remainder. The global average level of energy demand per capita increases marginally in the New Policies Scenario, from 1.9 tonnes of oil equivalent (toe) in 2011 to 2.0ථtoe in 2035. The large gap in energy demand per capita between OECD and non-OECD countries narrows over the projection period, but remains significant: in 2035, the OECD average is more than twoand-a-half times the non-OECD average. Comparisons at the extremes are even starker, with average per-capita energy demand in Africa being one-tenth or less of the levels in countries such as Canada, Russia and the United States in 2035.

Outlook by fuel

© OECD/IEA, 2013

Global demand for oil increases from 86.7ථmbͬd in 2011 to reach 101.4ථmbͬd in 2035. The average pace of demand growth slows over the period, from around 1.1% per year to 2020 to just 0.4% per year thereaŌer. Oil conƟnues to be the largest single component of the primary energy mix, but its share declines from 31% to 27%. While global oil demand grows, the overall change is the net result of decreasing demand in many OECD markets and increasing demand in many non-OECD markets, parƟcularly in Asia (where markets oŌen lack strong fuel-economy standards for vehicles) and the Middle Eastථ(where fossilfuel subsidies persist) (Figureථ2.4). The combinaƟon of rapidly increasing oil demand in China and decreasing demand in the United States (aŌer 2020), results in China overtaking the United States as the world’s largest oil consumer around 2030. Total oil demand growth in developing Asia is 13.9ථmbͬd to 2035, with India becoming the largest single source of growth aŌer 2020. Another pivotal development is the emergence of the Middle East as a major energy consumer, which, in the case of oil, results in its demand increasing by half to 2035 (reaching 9.9ථmbͬd), surpassing oil demand in the European Union before 2030. Oil demand is concentrated increasingly in the transport sector, which accounts for nearly 60%ථ(59ථmbͬd) of global oil demand in 2035. Fuel for road freight in Asia alone accounts, in energy terms, for one-third of the net global growth in oil demand over the Outlook period. Oil demand from road freight grows faster than that for passenger vehicles, increasing the weight of diesel in the overall road-transport fuel mix, which reaches 21ථmbͬd in 2035, geƫng close to the levels for gasoline (see Chapterථ15). Non-energy use – fuels used for feedstocks and non-energy products, such as asphalt, bitumen and lubricants – grows to 24ථmbͬd globally in 2035, about 70% of which is petrochemicals feedstocks. Global oil demand in industry remains broadly Ňatථin the New Policies Scenario (around 6.5ථmbͬd), while its use in power generaƟon halves and in buildings it falls by around 10%.

Chapter 2 | Global energy trends to 2035

61

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18

Figure 2.4 ‫ ٲ‬Change in oil demand in selected regions in the New Policies Scenario China

2011-2020

India

2020-2035

Middle East ASEAN Africa Brazil Other developing Asia Russia Japan United States European Union -4

-2

0

2

4

6

8 mb/d

Very strong growth in coal use over the last decade has resulted in the gap between primary energy demand for coal and oil narrowing signiĮcantlyථ(Figureථ2.5). In the New Policies Scenario, two-thirds of the projected growth in coal demand occurs before 2020, demand thereaŌer rising more gradually, to reach around 6ථ300ථmillion tonnes of coal equivalentථ(Mtce) inථ2035ථ(see Chapterථ4). Nearly three-quarters of this increase comes from the power sector. Coal conƟnues to be the largest source of fuel for power generaƟon, but its share declines from 47% in 2011 to 39% in 2035. China is by far the world’s largest coal producer and consumer and (as of 2012) the largest coal importer as well, having overtaken Japan. The growth in coal demand in China through to 2020 exceeds the growth in the rest of the world combined. However, this comparison masks a slowdown in coal demand growth in China that culminates in demand reaching a plateau before 2030. The scale of China’s coal use means that variaƟons in its demand for imports could have a big impact on the global picture. India becomes both the second-largest coal consumer – surpassing the United States – and the largest importer by 2025.

© OECD/IEA, 2013

Coal demand, similar to oil, declines in most OECD countries over the Outlook period, largely as a result of policies to reduce energy-related CO2 emissions from the power sector. Coal use in the United States declines by 14%, while, despite price dynamics currently supporƟng coal use, demand in the European Union falls by half byථ2035. Industry (including coking ovens and blast furnaces) dominates coal consumpƟon in end-use sectors and accounts for around one-quarter of global demand over the Outlook period, with iron and steel making up about half of this. Its use in industry increases by around 1.6% per year this decade, but then starts to decline. Global coal demand in buildings4 starts to decline this decade, while coal for non-energy use (such as petrochemical feedstock) becomes increasingly material, nearly tripling and overtaking use in buildings before 2030. 4.ഩ The buildings sector includes energy used in residential, commercial and institutional buildings, and nonspecified other. Building energy use includes space heating and cooling, water heating, lighting, appliances and cooking equipment. 62

World Energy Outlook 2013 | Global Energy Trends

In the New Policies Scenario, the absolute growth in primary demand for natural gas outpaces that of any other individual fuel (see Chapterථ3), and increases by more than the growth in demand for oil and coal combined from 2011 to 2035. Demand grows strongly throughout the Outlook period, and ends up nearly 50% higher, at 5ථtrillion cubic metresථ(tcm). Despite this strong growth, demand for natural gas remains below that for both oil and coal in 2035. Regional market dynamics conƟnue to be important, with gas prices reŇecƟng diīering gas supply and demand fundamentals, the nature of prevailing coal-to-gas compeƟƟon (see Chapterථ5) and the diīerent contract structures adopted. In the United States, gas demand increases relaƟvely slowly over the period – by 13% (over 90ථbillion cubic metres ΀bcm΁) – but it conƟnues to be the world’s largest gas market inථ2035. Demand in the European Union is also 13% higher in 2035, leaving it around 65ථbcmථ(10%) lower than projected in tKͲϮϬϭϮ͘This is, in part, due to a lower starƟng point, but also to a combinaƟon of factors that include more modest economic growth, increased eĸciency in buildings and the faster growth of renewables in power generaƟon.

Mtoe

Figure 2.5 ‫ ٲ‬World primary energy demand by fuel in the New Policies Scenario 5 000

2 3 4 5 6 7 8

Oil Coal Gas

4 000

1

9 3 000

10

2 000

Biomass

© OECD/IEA, 2013

1980

11

Nuclear Other renewables Hydro

1 000

1990

2000

2010

2020

12

2030 2035

Non-OECD countries account for more than 80% of global gas demand growth over the period to 2035. Demand for gas in developing Asia grows by around 680ථbcm, equivalent to the total amount of gas traded inter-regionally today. Demand grows quickly in Chinaථ(nearly 400ථbcm), but also briskly in Indiaථ(over 110ථbcm), Indonesiaථ(40 bcm) and other parts of the region. In absolute terms, demand for gas in the Middle East increases by more than the growth of the enƟre OECD – around 300ථbcm – between 2011 and 2035, driven by new power generaƟon (where demand for gas nearly doubles to reach 275ථbcm), desalinaƟon and higher industrial acƟvity. OŌen thought of primarily as an energy exporter, the Middle East increases its own natural gas use so rapidly that it overtakes the European Union before 2020 and consumes 26% more than the European Union by 2035. Russia, the world’s second-largest gas consumer, sees demand grow slowly (0.6% per year) as improved eĸciency and a move towards more market-based pricing help restrain demand growth. Gas demand in LaƟn America increases by around 85%, led by a 60ථbcm increase in Brazil as a result of the increased availability of domesƟc supplies (see Chapterථ10). Chapter 2 | Global energy trends to 2035

63

13 14 15 16 17 18

In the New Policies Scenario, power generaƟon conƟnues to be the largest source of gas demand, accounƟng for around 40% of global demand over the period. Around one-quarter of the net capacity addiƟons in the power sector between 2011 and 2035 are fuelled by natural gas (over 1ථ000ථgigawaƩs). Of the end-use sectors, industry sees the largest growth in gas demand in absolute terms (around 335ථbcm). Compared with tKͲϮϬϭϮ, gas use in industry in the United States is slightly higher in the Įrst half of the projecƟon period, but around the same level in 2035. This picture is subject to uncertainty, as several Įrms in energy-intensive industries have plans to relocate to North America to beneĮt from low gas pricesථ(see Chapter 8). In the European Union, industrial demand for gas declines by 10%, as a result of improvements in eĸciency and the conƟnuaƟon of a trend away from heavy industry to more light industry. China’s gas demand in industry increases by 14% per year to 2020 and reaches nearly 120ථbcm in 2035. Middle East demand for gas in industry overtakes that in the United States around 2030 and is around one-ĮŌh higher in 2035ථ(reaching 150ථbcm)͖ this is despite its economy being only around one-ĮŌh the size of the US economy at that Ɵme (in 2012 dollars at market exchange rates). Global gas demand in the buildings sector grows by 37%, driven by increased demand for space and water heaƟng, to reach around 985ථbcm in 2035. Natural gas use in transport doubles from 112ථbcm in 2011 to 225ථbcm in 2035, with a parƟcular focus on use in heavy-duty vehicles and Ňeet vehicles, such as buses and taxis.

© OECD/IEA, 2013

Nuclear power generaƟon increases by two-thirds in the New Policies Scenario, reaching 4ථ300ථterawaƩ-hours (TWh) in 2035. Demand is driven heavily by expansion in just a few countries: China accounts for around half of the global increase͖
World Energy Outlook 2013 | Global Energy Trends

projected to become the second-largest source of electricity before 2015 and approach coal as the primary source by 2035. China (mainly before 2020), India (mainly aŌer 2020), Brazil and Africa see noƟceable increases in hydropower. China, the European Union and the United States see the largest increases in electricity from wind and, by 2035, around 70% of the world’s wind power generaƟon capacity is expected to be in these three regions: 30% in China, 25% in the European Union and 14% in the United States. Prior to 2020, solar capacity addiƟons are concentrated in China, the European Union, Japan and the United States. AŌer 2020, solar capacity also increases rapidly in India and the MiddleථEast. Global demand for biofuels increases from 1.3ථmillion barrels of oil equivalent per dayථ(mboeͬd) in 2011 to 4.1ථmboeͬd in 2035. The share of biofuels in energy demand for road transport goes from 3% to 8%. The largest increases are seen in the United States, Brazil, European Union and China (but from a lower base).

The global energy map conƟnues to be transformed, with the weight of energy demand moving from OECD countries towards non-OECD countriesථ(Figureථ2.6). Non-OECD countries account for more than 90% of primary energy demand growth in the New Policies Scenario: more rapid populaƟon and economic growth, and increasing income, generates more demand for modern energy services. In 2004, the two groupings used about the same amount of energy but, by 2035, non-OECD demand is projected to be more than double that of OECD countries.

3 4 5

100%

60% 40%

1995

2005

2011

2015

2025

13 14 15

2035

In the New Policies Scenario, primary energy demand in the United States – the world’s second-largest energy consumer – increases to 2020 and then declines slightly to 2035. Over the Outlook period as a whole, US primary energy demand grows by around 2%. Oil demand in the United States in 2035 is around 20% lower than 2011 and only two-thirds of its historical peak in 2005. Demand for oil plateaus before 2020, at a level not much higher than today and, from that point, declines by around 3.7ථmbͬd to reach 14ථmbͬd in 2035. Fuel eĸciency standards play a major role in reducing gasoline demand, combined Chapter 2 | Global energy trends to 2035

9

12

OECD: United States Europe Japan Other

20%

8

11

Inter-regional Non-OECD: China India Brazil Middle East Other

80%

7

10

Figure 2.6 ‫ ٲ‬Share of world primary energy demand by region

© OECD/IEA, 2013

2

6

Regional trends

1985

1

65

16 17 18

with increasing use of alternaƟve fuels in transport, and there is a conƟnuing decline of oil use in most other sectors. Coal demand declines by 14% over the period, mainly as a result of policies to encourage a move towards other forms of power generaƟon and a reducƟon in use in industry. Helped by favourable prices and policies, natural gas demand increases by more than 90ථbcmථ(13%) through to 2035, with power generaƟon (60% of the increase), buildings and transport being the key growth sectors. Electricity generaƟon from renewables more than doubles, and accounts for around 23% of total generaƟon in 2035. Supported by producƟon tax credits, electricity generated from wind increases by 5% per year on average, and overtakes hydropower to become the largest source of renewablesbased generaƟon around the mid-2020s. Biofuels demand in the United States increases from less than 0.7ථmboeͬd in 2011 to 1.5ථmboeͬd in 2035, at the expense of oil products. Overall, shiŌs in energy demand and domesƟc supply (see energy supply secƟon) push the United States to the brink of being energy self-suĸcient in net terms in 2035: exports of coal and gas almost completely oīseƫng (in energy equivalent terms) the declining net imports of oil. Primary energy demand in the European Union declines by around 7% between 2011 and 2035. Demand for oil drops by one-third (3.7ථmbͬd). Gasoline and diesel each see a reducƟon of around 1ථmbͬd, as strict fuel-economy standards result in reduced demand in transport and the use of oil products in the buildings sector declines. Coal consumpƟon is half today’s level by 2035, falling by more than 200ථMtce, almost all of which is steam and brown coal use in the power sector. It takes around two decades for natural gas demand to get back to 2010 levels, with increases in the power sector and in buildings (where oil and coal use falls), but a decline in industry. Renewables increase their share of electricity generaƟon from 21% in 2011 to 44% in 2035, backed by renewables targets and ongoing support in the form of subsidies. GeneraƟon from wind grows parƟcularly strongly and it becomes the largest source of renewables-based generaƟon around 2020.

© OECD/IEA, 2013

Japan sees primary energy demand decline by 4% over the projecƟon period, with a reducƟon in energy use in transport and industry outweighing a slight increase in buildings. Oil consumpƟon declines by 36%, to less than 3ථmbͬd by 2035. Gas demand increases in end-uses, mainly buildings and industry. Electricity generaƟon from fossil fuels declines by around 110ථTWhථ(13%) over the projecƟon period, but this masks a signiĮcant decline in oil, a smaller decline in coal, and an increase in gas. Renewables-based generaƟon increases by 210ථTWh, accounƟng for 28% of total generaƟon in 2035 (solar and wind growing strongly). While our projecƟons show nuclear power providing 14% of electricity generaƟon in 2035, this is an area of parƟcular uncertainty. Japan is currently working on a new energy plan to be released in late-2013 and the way the future role of the Japanese nuclear industry is shaped will have major implicaƟons for the future of the rest of the power sector as well. In the New Policies Scenario, developing Asia accounts for 63% of global energy demand growth from 2011 to 2035. China, only recently established as the world’s largest energy consumer, is projected to consume in 2035 about 80% more energy than the United States (the next largest consumer)ථ(Figureථ2.7). China’s energy demand per capita increases by

66

World Energy Outlook 2013 | Global Energy Trends

40% over the Outlook periodථ(to reach 2.8ථtoeͬcapita), acceleraƟng away from the global average and geƫng close to the level of the European Union by 2035. In our projecƟons, China registers the largest energy demand growth in every major sector to 2035. Looking at growth by fuel is just as unambiguous, with China having a larger increase than any country in demand for oil, gas, nuclear, hydro, wind and solar. But the pace of energy demand growth does slow: growth this decade will be slower than the last, and in the 2020s growth will be less than half the level of the current decade.

global growth in the New Policies Scenario (Mtoe)

6

1 370

1 540 United States

5%

China 4 060

Middle East

2 240 1 050 1% Africa 480 Brazil 5%

1 030 8%

7 Japan

8

440

31%

0%* India

Southeast Asia

1 540

1 000

10%

9

11%

18%

10

Energy demand in 2035 %

3

5

E. Europe/Eurasia

0%*

2

4

Figure 2.7 ‫ ٲ‬Primary energy demand in selected regions and the share of

European Union

1

Share of net global energy demand growth (2011-35)

11

This map is without prejudice to the status of or sovereignty over any territory, to the delimitation of international frontiers and boundaries and to the name of any territory, city or area.

12

* These regions experience a decline in primary energy demand in 2035 relaƟve to 2011.

© OECD/IEA, 2013

China is about to become the world’s largest oil importer and becomes the world’s largest oil consumer around 2030 – reaching 15.1ථmbͬd in 2035. China’s road-transport Ňeet becomes the largest consumer of oil products of any Ňeet in the world around 2030 and, by 2035, consumes 7.9ථmbͬd. Policy acƟons to curb local polluƟon and meet energy security goals help to conƟnue the strong growth in gas and renewables and are an important factor in the slowdown in coal demand growth that is already occurring in China. The exact pace of this slowdown, and its impact on coal imports, conƟnues to be the biggest source of uncertainty for global coal markets. An increase in the naƟonal target for solar photovoltaics (PV) underpins a signiĮcant upward revision from tKͲϮϬϭϮ – with capacity reaching 35ථgigawaƩs (GW) in 2015 and nearly 160ථGW in 2035. China accounts for nearly 40% of world energy demand growth from 2011 to 2025, dominaƟng both the global and regional pictureථ(Figureථ2.8). AŌer 2025, the focus of demand growth shiŌs within developing Asia towards India and, to a lesser extent, Southeast Asia. In India, total primary energy demand more than doubles over the Outlook

Chapter 2 | Global energy trends to 2035

67

13 14 15 16 17 18

period͖ it all but matches that of the European Union in 2035, but on a per-capita basis it is sƟll only one-third of that level. India is projected to see the largest increase in coal demand globally, its consumpƟon doubling to reach around 970ථMtce in 2035. Oil demand in India reaches more than 8ථmbͬd in 2035, with road transport taking the largest share (a combinaƟon of growing vehicle ownership and relaƟvely low fuel eĸciency levels), but residenƟal demand for liqueĮed petroleum gas (LPG) and kerosene also accounts collecƟvely for nearly 1ථmbͬd. In our projecƟons, India meets the targets of its Solar Mission iniƟaƟve (22ථGW of capacity by 2022) and, assuming producƟon costs conƟnue to fall, is expected to have a relaƟvely large solar market by the 2020s. India increases its solar capacity by about 75ථGW between 2020 and 2035, second only to China and more than twice the increase in the European Union. Figure 2.8 ‫ ٲ‬Share of the growth in world primary energy demand by region in the New Policies Scenario 100%

3 847 Mtoe

2 806 Mtoe

1 511 Mtoe Rest of world Southeast Asia

80% 60%

India China 6% 10%

40%

9% 13% 14% 26%

20%

44%

37% 18%

© OECD/IEA, 2013

1995-2011

2011-2025

2025-2035

The importance of the Middle East as a centre of energy demand grows signiĮcantlyථ(Tableථ2.2). Total energy demand in the Middle East overtakes that of Russia around 2017 and is nearly 70% of the level of demand in the European Union in 2035. Oil demand in the Middle East grows to 10ථmbͬd in 2035, making it the world’s thirdlargest oil consumer (aŌer China and the United States). Gas consumpƟon eclipses that of the European Union by 2020 and reaches more than 700ථbcm by 2035 – second only to the United States. Gas demand in the power sector nearly doubles over the Outlook period. The share of renewables in electricity generaƟon increases from around 2.5% in 2011 to 13% in 2035. Energy demand for petrochemical feedstocks overtakes the level in the United States before 2030, the Middle East expanding its petrochemical producƟon substanƟally in 2035, consuming 2.2ථmbͬd of oil and 72ථbcm of gas as feedstock. By 2035, industry in the Middle East consumes 150ථbcm of natural gas – more than industry in China or the United States, where other fuels conƟnue to play a more signiĮcant role.

68

World Energy Outlook 2013 | Global Energy Trends

Table 2.2 ‫ ٲ‬World primary energy demand by region in the New Policies

1

Scenario (Mtoe) 1990

2000

2011

2020

2030

2035

2011-2035*

OECD



4 522

5 292

5 304

5 486

5 457

5 465

0.1%

Americas

2 260

2 696

2 663

2 811

2 826

2 850

0.3%

1 915

2 270

2 189

2 281

2 246

2 242

0.1%

1 630

1 765

1 778

1 763

1 719

1 709

-0.2%

631

832

863

912

912

906

0.2%

439

519

461

470

450

443

-0.2%

United States Europe Asia Oceania Japan Non-OECD

4 047

4 507

7 406

9 136

10 709

11 435

1.8%

E. EuropeͬEurasia

1 539

1 006

1 159

1 228

1 318

1 373

0.7% 0.7%

Russia

880

620

718

755

806

841

1 578

2 220

4 324

5 548

6 584

7 045

2.1%

China

879

1 175

2 743

3 519

3 945

4 060

1.6%

India

317

457

750

971

1 336

1 539

3.0%

Southeast Asia

223

373

549

718

897

1 004

2.5%

Asia

Middle East

212

358

640

796

970

1 051

2.1%

Africa

388

494

698

836

962

1 026

1.6%

LaƟn America

331

429

586

729

876

941

2.0%

138

184

267

352

441

480

2.5%

World**

8 769

10 071

13 070

15 025

16 623

17 387

1.2%

European Union

1 642

1 691

1 659

1 614

1 556

1 541

-0.3%

Brazil

3 4 5 6 7 8 9

*ථCompound average annual growth rate. ** World includes internaƟonal marine and aviaƟon bunkers (not included in regional totals).

Economic growth and a burgeoning middle class see energy demand in Brazil increase by about 80%, underlining its posiƟon as the dominant consumer in LaƟn Americaථ(see PartථB for an in-depth analysis of Brazil). Renewables conƟnue to meet a large part of total demand, with hydropower at the core of electricity supply and supply from bioenergy and wind increasing. Biomass also plays a signiĮcant role in industry in Brazil, while demand for biofuels in transport reaches 0.8ථmboeͬd in 2035, helping to slow oil demand growth.

© OECD/IEA, 2013

10 11 12 13 14

Sectoral trends In the New Policies Scenario, over half of the projected increase in global primary energy demand comes from the power sector – the result of conƟnuing electriĮcaƟon of the world economyථ(Figureථ2.9). Electricity demand expands most in buildings (in absolute terms), as a result of increased ownership of appliances and cooling needs in residences coupled with growing demand in the services sector (such as shops, oĸces, hotels and hospitals). The global average eĸciency of fossil fuel conversion in power plants improves by about 15%, but demand for energy inputs to generaƟon sƟll increases by 45%. Carbon capture and storageථ(CCS) technology appears well-suited to resolving at least some of the tension between rapidly increasing electricity demand, readily available exisƟng fossil

Chapter 2 | Global energy trends to 2035

2

69

15 16 17 18

fuel resources (and related infrastructure), and the need to limit CO2 emissions and local polluƟon, yet many signiĮcant challenges have sƟll to be overcome. They include the need to integrate component technologies eīecƟvely into large-scale projects, idenƟfy viable storage sites and put the necessary Įnancial incenƟves in place.6 At present, the outlook for CCS does not look bright and our projecƟons show only 67ථGW of CCS capacity in the power sector in 2035, around 1% of global fossil-fuelled power generaƟon capacity. The CCS capacity that does exist in 2035 is located mostly in the United States, China and the European Union. Figure 2.9 ‫ ٲ‬Change in energy demand by sector and fuel in the New

Mtoe

Policies Scenario, 2011-2035 2 500

Oil Coal

2 000

Gas Nuclear

1 500

Tradional biomass 1 000

Modern renewables Electricity and heat

500 0 -500

Power generaon and heat

Industry and nonenergy use

Transport

Buildings

Other*

© OECD/IEA, 2013

*ථIncludes other energy sector and agriculture.

Demand grows quickly – 1.7% per year – for those fuels that are used as a raw material for other products, mostly in the form of petrochemical feedstocks. The Middle East sees a signiĮcant increase in demand to 2035, with the availability of relaƟvely cheap feedstocks underlying a doubling of its petrochemical capacity over the projecƟon period. Emerging petrochemical producers in Asia, parƟcularly in China, Southeast Asia and India, also see substanƟally higher oil feedstock consumpƟon, driven by a rapidly increasing demand for plasƟcs. Globally, industrial energy use expands at 1.4% per year. China accounts for almost half of the growth to 2020, but its demand levels oī thereaŌer, when India, Southeast Asia and the Middle East account for much of the increase. Among the OECD regions, only North America sees any notable increase in industrial energy use, thanks in part to the boost to compeƟƟveness provided by relaƟvely low electricity and gas pricesථ(see Chapterථ8). In aggregate, OECD industrial energy demand grows modestly to 2020 and then levels-oī. Electricity and gas account for more than two-thirds of the demand increase from nonenergy intensive industries in the OECD. Despite the growth in energy demand for such industries, their share of total industrial energy demand grows only slightly. 6.ഩ For more on CCS, see the IEA’s dĞĐŚŶŽůŽŐLJZŽĂĚŵĂƉ͗ĂƌďŽŶĂƉƚƵƌĞĂŶĚ^ƚŽƌĂŐĞ(IEA, 2013b). 70

World Energy Outlook 2013 | Global Energy Trends

Global energy demand in transport grows at an average rate of 1.3% per year over the projecƟon period – a signiĮcantly lower rate of growth than seen in recent decades. All of the net growth comes from non-OECD regions, notably developing Asia͖ demand declines in the OECD, where eĸciency gains more than outweigh a modest expansion of the vehicle Ňeet. Although the number of cars and trucks on the world’s roads will roughly double between 2011 and 2035, advances in automoƟve technology lead to major improvements in average vehicle fuel economy. Globally, demand for diesel in road transport increases by 6.4ථmbͬd from 2011 to 2035, compared with a 2.1ථmbͬd increase in demand for gasolineථ(see Chapterථ15). Oil-based fuels conƟnue to dominate transport energy demand, though biofuels, and, to a much lesser extent, electricity for plug-in hybrid and electric vehicles account for a rising share of road-transport fuel demand. The use of natural gas in liqueĮed or compressed form grows rapidly, but from a small base (reaching 5.6% of total energy demand in transport in 2035 and 4.8% in road transport). United States and China lead the contribuƟon to growth, with low natural gas prices in the United States expected to push gas use in heavy trucks. However, while the technology is well-proven, the market remains small or non-existent in most countries, because of the obstacles to its adopƟon as a road fuel, for example, the lack of widespread refuelling infrastructure. AviaƟon and shipping become more fuel-eĸcient, oīseƫng to a large degree the eīect on fuel demand of the projected rise in demand for air travel and mariƟme freight. In the buildings sector, energy use grows at an average rate of 1% per year on average across the Outlook period, with nearly 75% of the growth coming from non-OECD countries. Households account for almost 60% of the increase in energy demand. Close to 1.8ථbillion new urban ciƟzens (mainly in developing countries) push up residenƟal demand, mostly in the form of electricity, because of strong growth in the use of appliances, space cooling and lighƟng. While the size of the world’s rural populaƟon remains stable (in absolute terms), and policies encourage a shiŌ to more eĸcient cookstoves, this only helps to limit growth in the use of biomass for cooking over the projecƟon period.

Energy resources The energy resources remaining in the world will not constrain the projected growth in energy demand to 2035 and well beyond. However, the scale of investment required to exploit them is huge and there are many factors that will determine the exact pace at which diīering energy resources will be developed, such as uncertainty around the economic outlook, the investment climate and availability of Įnancing, prevailing geopoliƟcs, energy and climate change policies, depleƟon policies in key producing regions, advances in technology and changes to legal, Įscal and regulatory regimes.7 © OECD/IEA, 2013

2 3 4 5 6 7 8 9 10 11 12 13

Energy supply

7.ഩ A WEO special report analysing the investment and financing needs of the world’s energy infrastructure will be published in mid-2014.

Chapter 2 | Global energy trends to 2035

1

71

14 15 16 17 18

High oil prices in recent years have supported an increase in total proven oil reserves, which are now esƟmated to be around 1ථ700ථbillion barrels, equivalent to 54ථyears of current producƟonථ(Figureථ2.10). Remaining recoverable resources are much larger: around 2ථ670ථbillion barrels of convenƟonal oil, 1ථ880ථbillion barrels of extra-heavy oil and bitumen, 1ථ070 billion barrels of kerogen oilථand 345ථbillion barrels of light Ɵght oilථ(LTO) (see Chapterථ13). Nearly 60% of remaining recoverable oil reserves are located onshore, 37% are oīshore (of which, more than one-third are in deepwater) and the remainder are in the ArcƟc. EsƟmates of remaining recoverable resources of oil conƟnue to increase as new technologies, such as mulƟ-stage hydraulic fracturing, unlock types of resources (such as LTO) that were not considered recoverable only a few years ago. Enhanced oil recoveryථ(EOR) technologies are currently esƟmated to have the potenƟal to unlock another 300ථbillion barrels from convenƟonal reservoirs (not included in our resource esƟmates) by increasing recovery rates, but realising the full potenƟal of EOR may be hampered in pracƟce by the complexity of EOR projects and shortage of the necessary skills in the industry. Figure 2.10 ‫ ٲ‬Fossil energy resources by type Total remaining recoverable resources Proven reserves Cumulave producon to date 3 050 years

233 years

178 years

142 years

Coal

61 years

54 years

Natural gas

Oil

© OECD/IEA, 2013

Notes: All bubbles are expressed as a number of years of producƟon based on esƟmated producƟon in 2013. The size of the bubble for total remaining recoverable resources of coal is illustraƟve and is not proporƟonal to the others. The Įgure speciĮes the status of reserves for coal as of end-2011, and gas and oil as of end2012. Sources: BGR (2012)͖ථOΘGJ (2012)͖ USGS (2000, 2012a and 2012b)͖ IEA esƟmates and analysis.

There are abundant proven reserves of coal – bigger than those for oil and gas combined in energy terms. These proven reserves increased by more than 3% in 2011 to reach an esƟmated 1ථ040 billion tonnesථ(BGR, 2012), equal to 142ථyears of producƟon at current ratesථ(see Chapter 4). Total remaining recoverable resources of coal are more than twenty Ɵmes the size of proven reserves and could support current producƟon levels for much longer. Both coal reserves and resources are distributed relaƟvely widely. Reserve levels are obviously far larger than needed to meet projected demand to 2035 and well in excess of the maximum which could be consumed without overshooƟng a 2ථΣC climate target (unless the CO2 emissions are miƟgated, such as by being captured and stored). 72

World Energy Outlook 2013 | Global Energy Trends

Proven resources of natural gas (both convenƟonal and unconvenƟonal) are esƟmated to be 211ථtcm, enough to sustain current levels of producƟon for 61ථyears. Remaining recoverable resources are assessed to be 810ථtcm and are equivalent to 233ථyears of producƟon at current ratesථ(see Chapterථ3). This assessment takes into account the latest esƟmate of shale gas resources from the USථEnergy InformaƟon AdministraƟon, which, mainly because it has broader coverage, is 10% higher than previously (US EIA, 2013). There are very large renewable energy resources – including bioenergy, hydro, geothermal, wind, solar and marine energy – which, if all harnessed, could meet projected energy demand many Ɵmes over. These resources are also very well spread geographically, relaƟve to other energy resources. However, in a number of cases, the cost of exploiƟng them on a large scale is currently prohibiƟve, even with government support. The potenƟal for renewables producƟon on an economically sound basis depends on how fast producƟon costs can be reduced: such cost reducƟons are already happening rapidlyථ(see Chapterථ6). Similarly, resources of uranium – the raw material for nuclear fuel – are more than adequate to supply the projected growth in nuclear power capacity through to 2035 and well beyond. Uranium resources expanded by 12.5% between the start of 2008 and 2011 and are suĸcient for over 100 years of supply, based on current requirements (NEAͬIAEA, 2012).

WƌŽĚƵĐƟŽŶŽƵƚůŽŽŬ

© OECD/IEA, 2013

In the New Policies Scenario, total oil producƟon8 increases by 11ථmbͬd from 2012 to reach 98ථmbͬd in 2035ථ(see Chapter 14). ProducƟon of crude oil declines by 4ථmbͬd over the Outlook period and its share of total oil producƟon declines from around 80% to twothirds. In contrast, producƟon of natural gas liquidsථ(NGLs) increases by 5ථmbͬd, with its availability being driven by growth in gas producƟon. UnconvenƟonal oil producƟon triples to reach 15 mbͬd in 2035 and, while it remains concentrated in North America, world LTO producƟon reaches nearly 6ථmbͬd by the late 2020s and remains around 5.6ථmbͬd in 2035.

1 2 3 4 5 6 7 8 9 10 11

Over the next decade, much of the net increase in global oil demand is met by non-OPEC supply, parƟcularly LTO in North America (mainly the United States), Canadian oil sands and deepwater pre-salt oil in Brazilථ(Figureථ2.11). The United States becomes the largest oil producer in the world (crude plus NGLs) in 2015 and retains this status unƟl the beginning of the 2030s. Brazil alone delivers more than one-third of the net global growth in oil producƟonථ– growing by more than double the increase in the United States – and becomes a net exporter around 2015ථ(see Chapterථ11). Oil producƟon falls in several regions, with Russia, the European Union and China seeing the biggest declines. From around the mid-2020s, OPEC oil producƟon growth (mainly from the Middle East) meets all of the global growth in demand, as non-OPEC producƟon starts to decline gradually. Over the projecƟon period, Iraq is by far the biggest contributor to OPEC producƟon growth, accounƟng for two-thirds of the totalථ(although Saudi Arabia remains the largest producer). OPEC’s share of global producƟon declines slightly by 2020 (from 43% to 41%), before then increasing to reach 46% in 2035.

12

8.ഩ Total oil ͞supply͟ denotes production of conventional and unconventional oil and NGLs plus processing gains (oil supply reaches 101.4 mbͬd in 2035), while oil ͞production͟ (discussed here) excludes processing gains. Processing gains are the volume increase in supply that occurs during crude oil refining.

18

Chapter 2 | Global energy trends to 2035

73

13 14 15 16 17

Figure 2.11 ‫ ٲ‬Change in production by fuel in selected regions in the New

mb/d

Policies Scenario 8

Oil: 2020-2035 2012-2020

6 4 2 0

Mtce

-2

Russia

China Venezuela United States

Caspian

Canada

Brazil

Middle East

300

Coal: 2020-2035 2011-2020

200 100 0

-100

bcm

-200

European United Union States

Africa

Australia Lan America

China

Indonesia

India

400

Natural gas: 2020-2035 2011-2020

300 200 100 0

Mtoe

-100

Brazil

Australia Caspian

Russia

United States

China

Africa

Middle East

200

Renewables: 2020-2035 2011-2020

150 100

© OECD/IEA, 2013

50 Middle East

Japan

ASEAN

India

Brazil

European Union

China

United States

Note: The change in producƟon through to 2035 is the summaƟon of the two periods shown in the Įgure.

74

World Energy Outlook 2013 | Global Energy Trends

© OECD/IEA, 2013

The global reĮnery sector is facing huge challenges: the changing composiƟon of feedstocks, changing product demand and the geographical shiŌ of demand away from OECD countries and towards Asia and the Middle Eastථ(see Chapterථ16). There is both a growing share in overall supply of extra-heavy oil, which requires more complex technology to process, and of NGLs, biofuels and coalͬgas-to-liquids, many of which bypass reĮneries completely. Rising demand for middle disƟllates, parƟcularly diesel, pushes reĮners to enhance yields for these products. Overall, global demand for reĮned products grows by 10ථmbͬd through to 2035, much less than the anƟcipated growth in overall liquids demand (16.8ථmbͬd, including biofuels) and less than net reĮnery capacity addiƟonsථ(13.1ථmbͬd).

1

In the New Policies Scenario, global coal producƟon increases by 15% from 2011 to more than 6ථ300ථMtce in 2035. ProducƟon in the European Union declines by nearly 60% over the Outlook period, responding to lower regional demand, reduced producƟon subsidies in some countries and cost escalaƟon. Coal supply in the United States starts to decline gradually before 2020 and is around 15% lower in 2035. Australia sees strong producƟon growth to 2020 and a more gradual increase thereaŌer. China’s producƟon increases by 9% and it remains the biggest coal producer over the period, accounƟng for around 45% of global producƟon in 2035. However, producƟon peaks before 2030 and then declines marginally. The absolute growth in coal producƟon in India is the largest of any country over the projecƟon period, helping to meet domesƟc demand for power generaƟon. The majority of the increase occurs aŌer 2020, when it accounts for more than 70% of global coal producƟon growth. Indonesia achieves a more than 80% increase in coal producƟon, both to meet domesƟc demand and for export. It overtakes Australia to become the fourthlargest coal producer on an energy equivalent basis.

5

2 3 4

6 7 8 9 10

World natural gas producƟon grows by 47% to 5ථtcm in 2035, with unconvenƟonal gas, LNG and evolving contractual structures all playing a role in the emergence of new players and an increasingly diverse trade picture. China sees the largest growth in gas producƟon (nearly 215ථbcm), two-thirds of this growth coming aŌer 2020. It becomes the world’s third-largest gas producer before 2025 (overtaking Qatar) and the second-largest producer of unconvenƟonal gas (aŌer the United States) before 2030. ProjecƟons for Russian gas producƟon are lower than in tKͲϮϬϭϮ, not as a result of supply constraints, but mainly due to modest growth in domesƟc demand and weak European import needs this decade. Despite this, Russia’s gas output rises by around 135 bcm to 2035 (all aŌer 2020), much of which goes to meet Asian demand. Turkmenistan sees producƟon double and its exports to China grow as the capacity of the Central-Asia pipeline is increased.

11

Currently an importer of both pipeline gas and LNG, Brazil increases gas producƟon by 7% per year on average, to reach more than 90ථbcm. The majority of this gas is associated gas from oil producƟon. It supports both a move to increasing gas use in power generaƟon, industry and buildings, and the aƩainment of self-suĸciency later in the projecƟon periodථ(see Chapterථ12). Elsewhere in LaƟn America, an assumed improvement in ArgenƟna’s investment climate facilitates a revival in gas producƟon, led by shale gas. The Middle East has more convenƟonal gas resources than any other region and sees its

16

Chapter 2 | Global energy trends to 2035

75

12 13 14 15

17 18

producƟon increase by more than 300ථbcm in the New Policies Scenario. Qatar, Iraq, Iran, Saudi Arabia and the United Arab Emirates all achieve signiĮcant increases in producƟon by 2035, but much of the gas goes toward meeƟng demand within the region. UnconvenƟonal gas is expected to account for nearly half of the global growth in producƟon over the Outlook period. However, the prospects for unconvenƟonal gas are parƟcularly uncertain, given the need to allay public concerns about the environmental and social implicaƟons and as-yet limited knowledge about the resource base in many parts of the world. As of 2012, the United States is the world’s largest gas producer (boosted by expanding supply of shale gas) and is expected to remain so through to 2035. Globally, producƟon of unconvenƟonal gas conƟnues to grow, reaching more than 830ථbcm by 2020 and more than 1ථ300ථbcm in 2035 – more than one-quarter of total gas producƟon at that Ɵme. North America leads the way, sƟll accounƟng for more than half of global unconvenƟonal gas producƟon in 2035, but the revoluƟon spreads aŌer 2020 and more than two-thirds of the supply growth over the projecƟon period as a whole occurs elsewhere (mainly China, Australia, India and ArgenƟna). Energy supply from renewables grows faster than any other source of energy, with two-thirds of the growth coming aŌer 2020. Most of the increase is supplied in the form of electricity, with wind and hydropower making the largest contribuƟon. In total, renewables-based generaƟon expands by more than two-and-a-half Ɵmes by 2035. The supply of bioenergy increases by over 40% over the Outlook period, with about half of the increase going to power generaƟon and much of the rest to the producƟon of biofuels (liquid road transport fuels). Biofuels producƟon grows from 1.3ථmboeͬd in 2012 to 4.1ථmboeͬd in 2035, with most of the increase coming from the United States and Brazil. While producƟon in the United States and the European Union is intended to meet domesƟc demand, Brazil is one of the few countries to develop producƟon capacity to serve other markets – Brazil’s net exports account for about 40% of global biofuels trade in 2035. China and India see producƟon increase aŌer 2020, but remain relaƟvely small compared with the United States and Brazil.

© OECD/IEA, 2013

Inter-regional energy trade The changes happening in the global energy system become strikingly evident when examining the projected future trends in inter-regional energy trade.9 Energy trade increases for all fossil fuels and for biofuels in the New Policies Scenario, with diīering, but profound, energy security and compeƟƟveness implicaƟons across regions. Oil remains the most heavily traded fuel, with trade increasing by around 5ථmbͬd to nearly 50ථmbͬd inථ2035. Overall, OECD net oil imports more than halve (to around 10ථmbͬd) and their share of total inter-regional trade declines from around 50% to only 20% in 2035. Light Ɵght oil (mainly through to 2020) and energy eĸciency (mainly aŌer 2020) combine to reduce US oil imports to around 3ථmbͬd in 2035. The added factor of producƟon from Canadian oil sands means that, collecƟvely, the United States and Canada become self-suĸcient in oil 9.ഩ Analysis is based on net energy trade between WEO regions. 76

World Energy Outlook 2013 | Global Energy Trends

before 2030. Combined, the two countries actually become energy self-suĸcient in net terms much sooner – around 2020.10 Imports into Europe also decline, but at a slower pace, and due to reduced demand. By contrast, Asia becomes the global centre of inter-regional crude oil trade – accounƟng for 63% of the world total in 2035.11 China is about to become the world’s largest oil importer, overtaking the United States, and goes on to surpass the oil imports of the European Union by 2020. By 2035, China’s oil imports reach 12.2ථmbͬd, geƫng close to the peak historical level of imports into the United States. India’s oil imports are larger than those of Japan by 2020 and exceed those of the European Union by 2035: its import dependence increases to more than 90%. Brazil undergoes a pivotal shiŌ, becoming a net oil exporter around 2015 and going on to export around 2.6ථmbͬd in 2035ථ(Figureථ2.12). Byථ2035, Southeast Asia will import around 60% more oil than the United Statesථ(over 5ථmbͬd). Exports from the Middle East are slightly lower than today in 2020, but then increase to reach 24.6ථmbͬd in 2035. The share of Middle East producƟon which is exported declines slightly, as domesƟc consumpƟon increases more quickly than producƟon. Russian oil exports decline to 6.2ථmbͬd, as new producƟon fails to keep pace with the decline in mature Įelds. Figure 2.12 ‫ ٲ‬Net oil and gas import/export shares in selected regions in the

Gas

New Policies Scenario 100% 80%

Net gas importer, net oil exporter

Japan and Korea

Net gas and oil importer

40%

2035

5 6 7 8

11 12

0%

Middle East

United States

-20%

Indonesia

Africa

Caspian

-60%

Net gas and oil exporter

-100% -100%

-80%

13

Southeast Asia

Russia

-40%

14

Net gas exporter, net oil importer -60%

-40%

-20%

0%

20%

40%

15 60%

80%

100% Oil

Notes: Import shares for each fuel are calculated as net imports divided by primary demand. Export shares are calculated as net exports divided by producƟon. A negaƟve number indicates net exports. Southeast Asia, ŝ͘Ğ͘ the ASEAN region, includes Indonesia. © OECD/IEA, 2013

4

India

20%

-80%

3

10

China

Brazil

2

9

2011

European Union

60%

1

17 18

10.ഩ Calculated on an energy-equivalent basis. 11.ഩ Developing Asia, Japan and
Chapter 2 | Global energy trends to 2035

16

77

Coal trade goes from 900ථMtce in 2011 to 1ථ260 Mtce in 2035, with most of the growth happening before 2020. As a share of total coal supply, trade increases from 16% in 2011 to 20% in 2035. China conƟnues to be the dominant coal importer for the remainder of this decade, with its imports increasing substanƟally to 2020, before then starƟng to decline. India overtakes China soon aŌer 2020 as the world’s largest coal importer. Despite its domesƟc resources, India’s coal imports more than triple over the Outlook period, reaching 350ථMtce in 2035 – more than one-quarter of global trade. Indonesia expands its coal exports by more than 50%, mainly in this decade, and Australian exports also grow by around half. The United States remains a signiĮcant net coal exporter throughout the period to 2035. Coal imports decline signiĮcantly in the European Union,
© OECD/IEA, 2013

Inter-regional natural gas trade increases by 2% per year, to reach nearly 1.1ථtcm in 2035. LNG accounts for nearly 60% of the increase in trade and, in combinaƟon with new sources of supply (convenƟonal and unconvenƟonal) and evolving contractual structures, boosts the Ňexibility of global gas supply. In general, exisƟng gas importers become more importdependentථ(the European Union, China, India), but there are notable excepƟons, such as the United States and Brazil. The United States moves to become a net exporter of gas in 2017 and is projected to export, in net terms, around 50ථbcm in 2035 as a result of new LNG export faciliƟes coming online.12 Like trade in oil, gas trade will also see its focus shiŌ to Asian markets, where the number of imporƟng countries will increase. In addiƟon, higher prices make Asia an aƩracƟve desƟnaƟon for many LNG cargoes. In many parts of the Middle East, there is a clear strategy to regard gas as a prime component of domesƟc supply and, in 2035, only 15% of Middle East gas producƟon is exported. Exports from Africa are projected to increase by 135ථbcm, as new producƟon in East Africa supplements supplies from other parts of the conƟnent. Gas exports from the Caspian region are projected to more than double and to go both east and west. LNG exports from Australia are projected to reach around 100ථbcm in 2035, while Russian exports increase by more than 65ථbcm and it remains the world’s largest gas exporter through to 2035. Regional gas price dynamics and evolving price mechanisms are important for gas markets over the Outlook period. In the New Policies Scenario, signiĮcant spreads between regional gas prices persist through to 2035, albeit with a limited degree of convergence (see Chapterථ1). Such diīerences in regional gas prices (together with electricity price diīerences) can aīect the compeƟƟveness of energy-intensive industries, such as chemicals, oil reĮning, iron and steel and others. In the New Policies Scenario, strong growth in demand for energy-intensive goods in many developing countries supports a swiŌ rise in their producƟon (and export expansion). RelaƟve energy costs play a more decisive role elsewhere, parƟcularly among OECD countries: natural gas and industrial electricity prices in the United States remain around half the level of the European Union and Japan in 2035. While the United States sees a slight increase in its share of global exports of energy-intensives goods, the European Union and Japan both see a strong decline in their export shares a combined loss of around one-third of their current shares (see Chapterථ8). 12.ഩ These net figures include Canadian exports by pipeline to the United States and US pipeline exports to Mexico. 78

World Energy Outlook 2013 | Global Energy Trends

Chapterථ3 also explores a case in which several factors combine to drive a much stronger convergence in regional gas prices towards a more global gas price (aŌer allowing for liquefacƟon and transport costs between regions). This case envisages increased linkages between regional markets and prices generally becoming more responsive to prevailing market condiƟons. In this Gas Price Convergence Case, global demand for gas is 107ථbcm higher in 2035 than the New Policies Scenario, with lower prices sƟmulaƟng demand in the European Union, Japan, China and other countries in Asia. Global gas trade is 5% higher in 2035, while gas import bills are lower in major gas-imporƟng regions, most notably China and the European Union.

1

Global trade in biofuels increases from 0.2ථmboeͬd in 2012 to 0.7ථmboeͬd in 2035. The United States remains the world’s largest producer, but becomes a net importer early in the projecƟon period (albeit with relaƟvely large imports and exports). Brazil is the main supplier to the internaƟonal market during the Outlook period and exports around 0.2ථmboeͬd by 2035 (see Chapterථ12), a signiĮcant porƟon of which goes to Europe.

5

/ŵƉůŝĐĂƟŽŶƐĨŽƌĞŶĞƌŐLJͲƌĞůĂƚĞĚK2 emissions It is extremely likely that human inŇuence has been the dominant cause of climate change since the mid-20th century, and very likely that it has contributed to observed global scale changes in the frequency and intensity of daily temperature extremes, according to the Intergovernmental Panel on Climate Changeථ(IPCC,ථ2013). As the source of more than twothirds of global greenhouse-gas emissions, the energy sector is crucial to tackling climate change. Global energy-related CO2 emissions in 2012 are esƟmated to be 31.5ථGt, a 1.2% increase over 2011. Over the past several decades, trends in emissions have followed those of the global economy closely, but they have shown increasing signs of divergence in more recent Ɵmes: one-quarter less CO2 is emiƩed today per unit of economic output than in 1990 (in PPP terms). The New Policies Scenario incorporates conƟnued support for renewables and eĸciency, an expansion of carbon pricing and a parƟal removal of fossilfuel subsidies. Even aŌer taking these factors into account, energy-related CO2 emissions increase by nearly 20%, to 37.2ථGt,ථin 2035. Nonetheless, there is an acceleraƟon in the divergence between emissions and economic growth, so that expanding the economy by one unit of GDP in 2035 emits nearly 50% less CO2 than similar economic expansion today.

© OECD/IEA, 2013

The New Policies Scenario points to an increase in the greenhouse-gas concentraƟon in the atmosphere, from 444ථparts per millionථ(ppm) in 2010 to over 700ථppm by 2100.13 This would correspond to an increase in the long-term global average temperature of 3.6ථΣC, compared with pre-industrial levels (an increase of 2.8ථΣC from today, adding to the 0.8ථΣC that has already occurred). By 2020, the level of emissions expected in the New Policies Scenario is already 3ථGt higher than under a trajectory compaƟble with limiƟng temperature increase to 2ථΣC, though addiƟonal correcƟve acƟon is sƟll possibleථ(Spotlight). 13.ഩ While the concentration of greenhouse gases measured under the
Chapter 2 | Global energy trends to 2035

79

2 3 4

6 7 8 9 10 11 12 13 14 15 16 17 18

S P O T L I G H T Redrawing the energy-climate map Policies currently under discussion are insuĸcient to limit the long-term global temperature increase to 2ථΣC, the target governments agreed at the United NaƟons Framework ConvenƟon on Climate Change Conference of the ParƟesථin Cancun, Mexico in 2010. The 450ථScenario demonstrates that reaching this target remains technically feasible, but intensive acƟon prior to 2020, the year in which a new internaƟonal climate agreement is due to come into force, is essenƟal. The WEO special report ZĞĚƌĂǁŝŶŐ ƚŚĞ ŶĞƌŐLJͲůŝŵĂƚĞ DĂƉ, published in June 2013, proposes a set of four fully economic policy measures that would, if implemented promptly, cut 80% of the excess emissions in 2020 relaƟve to the 2ථΣC target (IEA, 2013a). The four policy measures it set out in the 4-for-2ථΣC Scenario entail no net economic cost and would steer the world onto an emissions path that would keep the door open to achieving the 2ථΣC target. The policies were selected on the basis that they can deliver signiĮcant reducƟons in energy sector emissions by 2020 (as a bridge to further acƟon), rely only on exisƟng technologies, have already been proven in several countries, and their implementaƟon (as a package) would not harm economic growth in any region. The four policies are: „ AdopƟng speciĮc energy eĸciency measures (49% of the emissions savings). „ LimiƟng the construcƟon and use of the least-eĸcient coal-Įred power

plantsථ(21%). „ Minimising methane (CH4) emissions from upstream oil and gas producƟon (18%). „ AcceleraƟng the (parƟal) phase-out of subsidies to fossil-fuel consumpƟon (12%).

Targeted energy eĸciency measures would reduce global energy-related emissions by 1.5ථGt in 2020ථ(Figureථ2.13). These policies include imposing new or higher energy performance standards in many Įelds: in buildings, for lighƟng, new appliances and new heaƟng and cooling equipment͖ in industry, for motor systems͖ and, in transport, for road vehicles. Around 60% of the global savings in emissions are obtained in the buildings sector.

© OECD/IEA, 2013

Ensuring that new subcriƟcal coal-Įred plants are no longer built and limiƟng the use of the least eĸcient exisƟng ones would reduce CO2 emissions by 640 Mt in 2020 and also help curb local air polluƟon. Globally, the use of such plants would be one-quarter lower than would otherwise be expected in 2020. The largest emissions savings occur in China, the United States and India, all of which have a large number of coal plants. Methane (CH4) releases into the atmosphere from the upstream oil and gas industry would be almost halved in 2020, compared with the levels otherwise expected. Around 1.1 Gt carbon-dioxide equivalent (CO2-eq) of methane, a potent greenhouse gas, was released in 2010 by the upstream oil and gas industry. Reducing such releases into the 80

World Energy Outlook 2013 | Global Energy Trends

atmosphere represents an eīecƟve complementary strategy to the reducƟon of CO2 emissions. The necessary technologies are readily available, at relaƟvely low cost, and measures in this Įeld are being adopted in some countries, such as new performance standards in the United States. Accelerated acƟon towards a parƟal phase-out of fossil-fuel subsidies would reduce CO2 emissions by 360 Mt in 2020. Globally, fossil-fuel subsidies amounted to Ψ544ථbillion inථ2012, more than Įve Ɵmes the level of support to renewables. Figure 2.13ථ‫ ٲ‬Change in world energy-related CO2 emissions by policy

31

8

Fossil-fuel subsidies

33

9

29 27

NPS 2010

Policy measures 2020

4-for-2 °C

450S 2020

10

Note: NPS с New Policies Scenario͖ 450S с 450ථScenario. Source: IEA (2013a).

© OECD/IEA, 2013

IniƟaƟves and announcements since the publicaƟon of this WEO special report suggest that policymakers are giving close aƩenƟon to these four policy areas. The United States and China have signed an agreement to co-operate in combaƟng climate change, including by raising eĸciency in the transport and power sectors. The US President’s Climate AcƟon Plan includes strong acƟon across these policy areas. The Major Economies Forum has a new iniƟaƟve to improve the eĸciency of buildings. The investor community is moving towards more ambiƟous investment in low-carbon assets, while the World Bank will now provide Įnance to greenĮeld coal power projects only in rare circumstances. Other mulƟlateral investment banks are also considering adopƟng this posiƟon. The very long lifeƟme of some greenhouse-gases means that their cumulaƟve build-up in the atmosphere is an important consideraƟon. The IPCC concludes that the world has a maximum global CO2 emissions budget of 1ථ133ථGt from 2012 onwards if it is to keep to a 50% chance of limiƟng the long-term average increase in global temperature to no more than 2ථΣCථ(IPCC, 2013).14 Based on this esƟmate and the New Policies Scenario, 74% of the available CO2 emissions budget will be consumed by the energy sector alone by 2035. If 14.ഩ This IPCC estimate takes account of radiative forcing from other sources.

Chapter 2 | Global energy trends to 2035

4

7

Energy-related CO2 Energy efficiency

Upstream CH4 reducons

35

3

6

Upstream oil and gas CH4 emissions Power generaon

Gt CO2-eq

Other energyrelated CH4

37

2

5

measure in the 4-for-2 °C Scenario 39

1

81

11 12 13 14 15 16 17 18

unabated, the potenƟal CO2 emissions from consuming all fossil fuel reserves (as of 2012) would amount to 2ථ860ථGt – two-and-a-half Ɵmes the IPCC’s esƟmate of the maximum global CO2 emissions budget. This is before factoring in remaining recoverable fossil fuel resources (which are much larger than proven reserves) or non-energy related CO2 emissions, such as deforestaƟon. Such analysis puts into sharp focus the need to increase the adopƟon of technologies such as CCS rapidly and at scale if the world is to balance use of its fossil fuel resources with meeƟng its environmental objecƟves. The geographical distribuƟon of energy-related CO2 emissions is set to change signiĮcantly between now and 2035. All of the growth occurs in developing countries, as emissions across the OECD declines by 16%, to 10.2ථGt in 2035, due to saturaƟon of energy demand and the aīects of policies promoƟng energy eĸciency and decarbonisaƟon of the fuel mix. China is expected to remain the largest emiƩer throughout the projecƟon period. Chinese emissions are 60% larger than those of the United States in 2012, but will be more than twice the size of the United States by 2035ථ(Figureථ2.14). Emissions in India are expected to overtake those of the European Union in the mid-2020s and get closer to the levels of the United States in 2035. By the end of the projecƟon period, emissions from both Southeast Asia and the Middle East will be at a similar level to those of the European Union. Figure 2.14 ‫ ٲ‬Energy-related CO2 emissions by region in the New

Rest of world*

35

Japan

30

Southeast Asia

Gt

40

100% 80%

Middle East

25

60%

European Union

20

India

15

United States

10

China

40%

Share of global CO2 emissions

Policies Scenario

20%

5 1900 1920 1940 1960 1980 2000 2020 2035

1990 2011 2035

© OECD/IEA, 2013

*ථRest of world includes internaƟonal bunkers.

CO2 emissions per capita decline slightly on a global basis, to 4.3 tonnes per capita, in 2035. On average, OECD countries see their CO2 emissions per capita decline by nearly one-quarter. Per-capita emissions in the United States drop signiĮcantly, but remain nearly three Ɵmes the world average in 2035͖ those of Japan decline to 7.8ථtonnesͬcapita and the average level of the European Union falls to 4.9ථtonnesͬcapitaථ(Figureථ2.15). Some developing economies experience rapid increases in per-capita emissions: China converges to the OECD average in 2035, while the Middle East overtakes the OECD average. In Southeast Asia and India, per-capita emissions remain below the world average, despite an increasing trend over the projecƟon period. 82

World Energy Outlook 2013 | Global Energy Trends

Figure 2.15 ‫ ٲ‬Energy-related CO2 emissions per capita and CO2 intensity in

1

Tonnes per thousand dollars of GDP (2012, MER)

selected regions in the New Policies Scenario 1.2

2012

1.0 0.8

3

2035

India

China Russia

0.6

Middle East

0.4

Bubble area

5

2 Gt United States

European Union 0

3

Japan 6

9

12

4 Gt

6

15 18 Tonnes per capita

Emissions from coal remain the largest source of energy-related CO2 emissions throughout the period, but they stabilise in 2025 at around 15.7 Gt. More than 45% of the growth in global emissions from 2012 to 2035 is expected to come from natural gas, despite its lower level of emissions per unit of energy. By 2035, natural gas combusƟon releases over 9.1 Gt of CO2, while oil combusƟon accounts for 12.5ථGt. Emissions are expected to rise in all sectors over the Outlook period, with the largest increase in the power sector (2.1ථGt), driven by increasing electricity demand in developing countries, mostly in buildings. The power sector in China alone adds 1.3 Gt through to 2035, even though the share of nonfossil generaƟon expands from 22% in 2012 to 38% in 2035. Global CO2 emissions from the transport sector expand by 2.0ථGt, with developing Asia accounƟng for nearly threequarters of this growth. Expanding demand for mobility, oŌen coupled with subsidised prices and weak or non-existent fuel-economy standards, explains this growth.

Topics in focus

© OECD/IEA, 2013

4

1 Gt

Southeast Asia

0.2 0

2

Key

7 8 9 10 11 12 13

This secƟon presents new data and analysis on three topics that have an important bearing on the Outlook for the global energy system. The ten members of the AssociaƟon of Southeast Asian NaƟons (ASEAN) are, together with China and India, shiŌing the centre of gravity of the global energy system toward Asia. We examine in greater detail the current energy situaƟon in Southeast Asia and the important trends inŇuencing its energy outlook to 2035. tKͲϮϬϭϯ also conƟnues its coverage of the need to increase modern energy access to the huge number of people in the world currently without it, and the imperaƟve to phase out ineĸcient fossil-fuel subsidies that serve to distort energy markets. Here we present our latest data and analysis, as well as covering key developments over the last year and, in the case of energy access, our projecƟons for the future.

14 15 16 17 18

Chapter 2 | Global energy trends to 2035

83

Energy trends in Southeast Asiaϭϱ dŚĞƐŝƚƵĂƟŽŶƚŽĚĂLJ Since 1990, Southeast Asia’s energy demand has expanded two-and-a-half Ɵmes. By 2011, it had reached 550ථMtoe, or around three-quarters of that of India. Considerable further growth in demand can be expected in the region, especially considering that the per-capita energy use by its 600ථmillion inhabitants is low, at just half of the global average, and the region’s strong long-term economic growth prospects. However, the countries of Southeast Asia are extremely diverse, with vast diīerences in their scale and paƩern of energy use and their energy resource endowmentsථ(Figureථ2.16). Indonesia, the largest energy user in the region, with 36% of overall demand, consumes 66% more energy than Thailand (the second-largest user) and over 50ථƟmes more energy than Brunei Darussalam (which has the lowest consumpƟon and a much smaller populaƟon). Compared with some of its neighbours, Southeast Asia is richly endowed with fossil and renewable energy resources, though they are distributed unevenly across the region and oŌen located far from demand centres. Currently, the region is an exporter in net energy-equivalent terms, as exports of coal (220ථMtce), natural gas (62ථbcm) and biofuels more than oīset net imports of oil (1.9ථmbͬd). Indonesia is by far the dominant producer, having greatly increased its coal output and exports in the last decade. Phasing out fossil-fuel subsidies and providing access to modern energy services remain unĮnished business in Southeast Asia. Fossil-fuel subsidies amounted to Ψ51ථbillion in the region in 2012. Despite recent reform eīorts, notably in Indonesia, Malaysia and Thailand, subsidies remain a signiĮcant factor distorƟng energy markets. They encourage wasteful energy consumpƟon, burden government budgets and deter investment in energy infrastructure and eĸcient technologies (see Įnal secƟon of this chapter). With one-ĮŌh of the populaƟon in the region sƟll lacking access to electricity and almost half sƟll relying on the tradiƟonal use of biomass for cooking, much remains to be done to achieve universal access to modern energy. In Indonesia, for example, electricity demand was lower than Norway’s unƟl the mid-2000s, yet its populaƟon is some 50 Ɵmes greater.

© OECD/IEA, 2013

DĞĞƟŶŐĨƵƚƵƌĞŶĞĞĚƐ In the New Policies Scenario, Southeast Asia’s energy demand increases by over 80% between 2011 and 2035, a rise equivalent to current demand in Japan. This supports a near tripling of the region’s economic acƟvity and a populaƟon increase of almost onequarter. Oil demand rises from 4.4ථmbͬd today to 6.8ථmbͬd in 2035, almost one-ĮŌh of projected world growth. AŌer having grown at 10% per year, on average, since 1990, coal demand triples over the period to 2035, accounƟng for nearly 30% of global growth in coal use. Natural gas demand increases by around 80% to 250ථbcm. The share of renewables in the primary energy mix falls, even with rapidly increasing use of modern renewables – such as geothermal, hydropower and wind – and relaƟvely stable use of tradiƟonal biomass for cooking. 15.ഩ This section summarises the findings of a WEO special report presented at the 7th East Asia Summit Energy Ministers Meeting in Bali on 26 September 2013. 84

World Energy Outlook 2013 | Global Energy Trends

© OECD/IEA, 2013

Chapter 2 | Global energy trends to 2035

Figure 2.16ථ‫ ٲ‬Energy in Southeast Asia Myanmar

Lao PDR

Abundant hydropower and natural gas resources; their development is vital to reduce poverty and support economic growth.

Aims to become the hydropower “battery” of Asia; electricity exports have been increasing sharply.

Vietnam Significant renewable and fossil energy resources, but rapidly growing energy demand underlies a shift towards imports; developing a nuclear power programme.

Philippines Hanoi

Fast rising electricity demand requires expanded supplies; strongly reliant on energy imports, though it is the world's second-largest geothermal producer.

Thailand Vientiane

Second-largest energy consumer in ASEAN and heavily dependent on energy imports due to limited energy resources; aims to diversify electricity generation.

Yangon

Malaysia

Manila

Bangkok

Third-largest energy consumer in ASEAN with relatively high per-capita consumption; significant oil and LNG exporter, but production is maturing.

Phnom Penh

Cambodia Low levels of electrification, although improving; potential to develop oil and gas resources.

Brunei Darussalam Bandar Seri Begawan

Among the wealthiest countries in the world on a per-capita basis, thanks to oil and LNG exports.

Kuala Lumpur

Singapore Singapore

Strategically situated, it has become Asia's key oil trading and refining hub (the third-largest in the world) and could become a major gas hub.

Jakarta

Indonesia Largest energy consumer in ASEAN, with massive scope for growth; it exports steam coal (the world's largest) and LNG, and is an increasing importer of oil. 85

This map is without prejudice to the status of or sovereignty over any territory, to the delimitation of international frontiers and boundaries, and to the name of any territory, city or area.

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

The power sector is fundamental to the energy outlook for Southeast Asia and, within it, coal emerges as the fuel of choice. Electricity generaƟon between 2011 and 2035 increases by more than the current power output of India. Coal’s relaƟve abundance and aīordability in the region boosts its share of electricity generaƟon from less than one-third today to almost half in 2035, at the expense of natural gas and oil. This shiŌ is already underway: some three-quarters of the thermal capacity now under construcƟon is coalĮred. Deploying only more eĸcient coal-Įred power plants should be a major priority in the region – the average eĸciency is currently just 34%, owing to the almost exclusive use of subcriƟcal technologies. If the region’s coal-Įred power plants were as eĸcient as those in Japan today, their fuel use would be one-ĮŌh lower, and CO2 emissions and local air polluƟon be substainƟally reduced. Southeast Asia faces sharply increasing reliance on oil imports, which will impose high costs and leave the region more vulnerable to potenƟal disrupƟons. Decline in mature Įelds and the absence of large new prospects lead oil producƟon across the region to fall by almost one-third in the period to 2035. As a result, Southeast Asia becomes the world’s fourthlargest oil importer, behind China, India and the European Union. Its oil import dependency almost doubles, to 75%, as net imports rise from 1.9ථmbͬd to just over 5ථmbͬd. The region’s spending on net oil imports triples to almost $240ථbillion in 2035, equivalent to almost 4% of GDP. Spending in Thailand and Indonesia on net oil imports triples to nearly $70ථbillion each in 2035. There will be a reduced surplus of natural gas and coal from the region for export, as producƟon is increasingly dedicated to domesƟc markets. Despite increasing gas producƟon, Southeast Asia’s net gas exports, which come mainly from Indonesia, Malaysia, Myanmar and Brunei Darussalam, are projected to be cut from 62ථbcm to 14ථbcm in the period to 2035. The region’s net coal exports also decline aŌer 2020, as regional demand outpaces growth in indigenous producƟon, though Indonesia’s coal producƟon rises by more than 80%, to 550ථMtce in 2035 and it remains one of the world’s biggest coal producers and, by a very large margin, the top exporter of steam coal.

© OECD/IEA, 2013

Developing policies to aƩract investment will be vital for enhancing energy security, aīordability and sustainability. Around $1.7ථtrillion of cumulaƟve investment in energysupply infrastructure to 2035 is required in the region, with almost 60% of the total in the power sector. Mobilising this will be challenging unless acƟon is taken to eliminate exisƟng barriers, which include subsidised energy prices, under-developed energy transport networks, and instability and inconsistency in the applicaƟon of energy-related policies. While Southeast Asia has made some gains in energy eĸciency, almost three-quarters of the full economic potenƟal is set to remain untapped in 2035. Removing barriers to energy eĸciency deployment would, accordingly, deliver major energy savings, as demonstrated in the Eĸcient ASEAN Scenario of the WEO special report Southeast Asia Energy Outlook, which provides for the uptake of energy eĸciency opportuniƟes that are both economically viable and have acceptable payback periods (IEA,ථ2013c). Compared with the New Policies Scenario, energy demand is cut by almost 15% in 2035, an absolute amount that exceeds 86

World Energy Outlook 2013 | Global Energy Trends

Thailand’s current energy demand. Lower electricity demand and the use of more eĸcient power plants reduce coal demand by 25%. More eĸcient industrial equipment, stringent vehicle fuel-economy standards and the quicker phase-out of fossil-fuel subsidies drive demand reducƟons in oil (10%) and gasථ(11%). The region’s net oil imports are cut by around 700ථkbͬd in 2035, a level comparable to Malaysia’s current producƟon, cuƫng oilimport bills by $30ථbillion. By the end of the period, net exports of natural gas are three Ɵmes higher than in the New Policies Scenario (reaching 42ථbcm) and of coal 50% higher (reaching 320ථMtce). Unlocking Southeast Asia’s energy eĸciency potenƟal requires government acƟon to address a wide spectrum of barriers. The measures to be adopted will vary by country and by sector, but priority areas include vehicle fuel-economy standards, more stringent building codes and energy performance standards for a wider range of appliances and products. Improving administraƟve experƟse and energy data collecƟon are essenƟal pre-requisites to developing eīecƟve energy eĸciency policies and their implementaƟon. RealisƟc and measurable eĸciency targets are needed, along with selecƟve measures to achieve them and mechanisms to monitor progress and make adjustments as required. Energy eĸciency investments need to be made more aīordable, both by eliminaƟng market distorƟons and by increasing the availability of Įnancing and incenƟves. Carefully constructed packages could supply Įnancial support to those who need it most, drawing on funds released by the progressive eliminaƟon of consumer subsidies.

DŽĚĞƌŶĞŶĞƌŐLJĨŽƌĂůůϭϲ There is growing recogniƟon that modern energy is crucial to achieving a range of social and economic goals relaƟng to poverty, health, educaƟon, equality and environmental sustainability, and this recogniƟon is reŇected in a number of new iniƟaƟves. A United NaƟons High Level Panel of Eminent Persons has recommended that universal access to modern energy services be included in the Post-2015 Development Agenda. The United States has launched a Power Africa iniƟaƟve, aimed at doubling electricity access in subSaharan Africa over Įve years. At the Ɵme of wriƟng, 77ථdeveloping countries have signed up to the UNථSustainable Energy for All (SE4All) iniƟaƟve, including many of those with the largest populaƟons lacking access to modern energy. Many businesses, aid organisaƟons and non-governmental organisaƟons have also joined the SE4All iniƟaƟve.

© OECD/IEA, 2013

Alongside this increase in poliƟcal focus, the last year has seen new analysis which increases our understanding of energy access. The Įrst major analyƟcal report produced under the SE4All iniƟaƟve, 'ůŽďĂůdƌĂĐŬŝŶŐ&ƌĂŵĞǁŽƌŬ, which was led by the IEA and the World Bank, deĮnes the starƟng point against which progress can be measured and the scale of the challenge understood (IEA and World Bank, 2013). In addiƟon, new research Įnds that 16.ഩ In this analysis, we define access to modern energy services as household access to electricity and clean cooking facilities. It is recognised that this excludes some important categories, such as access to energy for productive use, for community services and for heating. While this is an imperfect situation, such categories are often excluded from quantitative analysis of energy access due to the lack of comprehensive, reliable data. See tKͲϮϬϭϮ and our energy access methodology for a fuller discussion of these issues, both available at ǁǁǁ͘ǁŽƌůĚĞŶĞƌŐLJŽƵƚůŽŽŬ͘ŽƌŐͬĞŶĞƌŐLJĚĞǀĞůŽƉŵĞŶƚ.

Chapter 2 | Global energy trends to 2035

87

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18

there are 3.5ථmillion premature deaths each year as a result of household air polluƟon from using solid fuels (rising to 4ථmillion, if the contribuƟon of household air polluƟon to outdoor air polluƟon is included). This Įgure is much higher than previous esƟmates, primarily due to the inclusion of new diseases, such as cardiovascular disease and lung cancer (Lim, ĞƚĂů͕͘ථ2012).

ƵƌƌĞŶƚƐƚĂƚƵƐŽĨĞŶĞƌŐLJĂĐĐĞƐƐ Modern energy for all is far from being achieved. We esƟmate that nearly 1.3 billion people, or 18% of the world populaƟon, did not have access to electricity in 2011 – 9ථmillion fewer than in the previous yearථ(Tableථ2.3).17 The global improvement since last year is modest, while the picture for some countries has worsened. Sub-Saharan Africa and developing Asia account collecƟvely for more than 95% of the global total. The populaƟon without access to electricity in sub-Saharan Africa is now almost equal to that of developing Asia and, if current trends conƟnue, will overtake it in the near future.18 Since 2000, around two-thirds of the people gaining access to electricity have been in urban areas and the populaƟon without electricity access has become more concentrated in rural areas.

© OECD/IEA, 2013

At a country level, the latest esƟmates conĮrm the progress that China and Brazil have made over many years in increasing access to electricity and that they are now geƫng close to the goal of universal electriĮcaƟon. In Asia, the latest esƟmates reveal improvements in electricity access in Bangladesh, Indonesia and Sri Lanka. India remains the country with the largest populaƟon without electricity access at 306 million people.19 Experience in Pakistan serves to highlight a diīerent element of the energy access challenge, that of achieving reliability of supply, as fuel shortages have jeopardised electricity supply there and resulted in prolonged load-sheddingථ(Boxථ2.2). In Africa, the latest esƟmates reveal improvements in South Africa, Ghana, Cameroon and Mozambique, all of which have explicit plans in place to boost electricity access. The Power Africa iniƟaƟve is supporƟng these eīorts, with the US government having commiƩed more than $7ථbillion, through a combinaƟon of loans, guarantees, credit enhancements and technical assistance. Private companies have agreed to put up an addiƟonal $9ථbillionථ(US Government, 2013). Partner countries already include Ethiopia, Ghana,
17.ഩ Our estimates are based on 2011 data where available or an estimate based on latest available data. 18.ഩ tKͲϮϬϭϰ will include a special focus on energy developments in Africa. 19.ഩ Our estimates for India are based on the latest National Sample Survey and are in line with those published in India’s 12th Five-zear Planථ(Planning Commission of India,ථ2013). However, the Five-zear Plan also notes that the 2011 Census of India reports a 67.2% national electrification rate, which is lower than the latest National Sample Survey. Applying the rate reported in the Census results in the estimated number of people in India without access to electricity increasing to around 410ථmillion in 2011, which would change our global estimate to around 1.4 billion. India’s 12th Five-zear Plan notes this difference in estimates, stating that it is possibly due to differences in questionnaire design and that it needs to be looked into further. 88

World Energy Outlook 2013 | Global Energy Trends

Table 2.3 ‫ ٲ‬Number of people without access to modern energy services by region, 2011 (million)20

Developing countries Africa Sub-Saharan Africa Nigeria South Africa North Africa Developing Asia

Without access to electricity Share of PopulaƟon populaƟon 1ථ257 23%

TradiƟonal use of biomass for cooking* Share of PopulaƟon populaƟon 2ථ642 49%

600

57%

696

67%

599

68%

695

79%

84

52%

122

75%

8

15%

6

13%

1

1%

1

1%

615

17%

1ථ869

51%

India**

306

25%

818

66%

Pakistan

55

31%

112

63%

Indonesia

66

27%

103

42%

3

0%

446

33%

24

5%

68

15%

1

1%

12

6%

19

9%

9

4%

1ථ258

18%

2ථ642

38%

China LaƟn America Brazil Middle East World***

2 3 4 5 6 7 8

* Based on World Health OrganizaƟon (WHO) and IEA databases. ** Since tKͲϮϬϭϮ, populaƟon numbers for India have undergone a signiĮcant upward revision (See Chapterථ1 for populaƟon assumpƟons), meaning that, while the electriĮcaƟon and clean cooking access rates have not changed, the number of people esƟmated to be without access has signiĮcantly increased. See also footnoteථ19. ***ථIncludes OECD countries and Eastern EuropeͬEurasia.

© OECD/IEA, 2013

We esƟmate that more than 2.6 billion people, or 38% of the global populaƟon, relied on the tradiƟonal use of biomass for cooking in 2011 – 54 million more people than in the previous year.21 This deterioraƟng situaƟon is primarily due to populaƟon growth outpacing improvements in the provision of clean cooking faciliƟes. The esƟmates reveal a worsening situaƟon in sub-Saharan countries such as Nigeria, Uganda,
Chapter 2 | Global energy trends to 2035

1

89

9 10 11 12 13 14 15 16 17 18

Box 2.2 ‫ ٲ‬Fuel shortages in Pakistan Pakistan faces economic and energy challenges that intersect most clearly in relaƟon to electricity supply. Around 55ථmillion people – more than 30% of the populaƟon – do not have access to electricity. Of those that do have electricity, the quality of supply they receive can be a major source of frustraƟon. While Pakistan has 23ථGW of installed power generaƟon capacity, the cost of fuel has proved to be a signiĮcant Įnancial burden to generators, relaƟve to the price they can charge for power, resulƟng in shortages and power cuts. The share of oil in the generaƟon mix is relaƟvely high and the doubling of electricity tariīs since 2008 has not been suĸcient to compensate for rising fuel costs. The problem is made worse by a long legacy of unpaid energy bills and distribuƟon losses (oŌen due to theŌ). State-owned power companies have faced large losses and accumulated debt that government subsidies are unable to cover fully. This has resulted in power companies being unable to buy suĸcient fuel, which, in turn, has prompted extensive load shedding – up to 12ථhours per day in urban areas and 20ථhours per day in rural areasථ(NEPRA, 2012). Such prolonged power shortages have a major impact on Pakistan’s economy, cuƫng GDP growth by an esƟmated 2%ථ(ADB, 2013).

© OECD/IEA, 2013

The Asian Development Bank is supporƟng government eīorts to increase power generaƟon capacity, improve transmission and distribuƟon, and deliver renewable energy projects. Pakistan has also recently agreed funding support from the government of Saudi Arabia to complete a 1ථGW hydro project (Arab News, 2013) and, in September 2013, reached agreement with the InternaƟonal Monetary Fund on a $6.7ථbillion loan, linked to energy sector reforms. In mid-2013, the government also took steps to help clear the debt of independent power producers. In the longer term, the power sector will need to be restructured, including the introducƟon of tariīs that fully reŇect underlying costs and beƩer revenue collecƟon and enforcement. Such reforms can be easier to implement as the quality of service improves. Several countries are taking acƟon to expand access to clean cooking faciliƟes. Indonesia has set a highly ambiƟous target of enabling 85% of households to use LPG or natural gas for cooking by 2015. The kerosene-to-LPG conversion programme implemented in 2007 has successfully decreased the use of kerosene, a relaƟvely polluƟng fuel, but the shiŌ from biomass to gas remains a challenge. While subsidies to LPG represent an important cost of transiƟon to clean fuels in Indonesia, they represent a net saving in cases where households are switching from kerosene, which receives higher subsidies. In Africa, Ghana’s government has commiƩed to the very ambiƟous goal of bringing LPG to half the number of households, more than doubling the current level. Nigeria, Africa’s most populous country, has set a naƟonal goal of helping 10ථmillion households (around one-third of the total) to switch to clean cooking faciliƟes by 2021͖ Nigerian households currently rely heavily on tradiƟonal biomass for cooking despite the country’s abundant fossil fuel resources. InternaƟonal eīorts are also being stepped up. The Global Alliance for Clean Cookstoves plans to promote the adopƟon of clean cookstoves and fuels to 100ථmillion households by 2020ථ(GACC, 2012). It has prioriƟsed acƟon in six countries: Bangladesh, 90

World Energy Outlook 2013 | Global Energy Trends

China, Ghana,
Outlook for energy access in the New Policies Scenario In the New Policies Scenario, the number of people without access to electricity is projected to decline by more than one-ĮŌh to around 970ථmillion in 2030, or 12% of the global populaƟon (Tableථ2.4).22 Around 1.7 billion people are expected to gain access over the period to 2030 but, in many cases, these gains are oīset by populaƟon growthථ(increases by 1.4ථbillion to 2030). While there is an improving global picture, the regional trends are very diverse. Developing Asia sees the number of people without access to electricity decline by around 290 million between 2011 and 2030. China is expected to achieve universal access within the next few years. India sees a signiĮcant improvement: its electriĮcaƟon rate rises from 75% today to around 90%, but the country sƟll has, in 2030, the largest number without access to electricity in any single country. Table 2.4 ‫ ٲ‬Number of people without access to modern energy services by region in the New Policies Scenario, 2011 and 2030 (million) Without access to electricity 2011 2030

Without access to clean cooking faciliƟes 2011 2030

1ථ257

969

2ථ642

2ථ524

600

645

696

881

599

645

695

879

615

324

1 869

1 582

China

3

0

446

241

India

306

147

818

730

24

0

68

53

Developing countries Africa Sub-Saharan Africa Developing Asia

LaƟn America Middle East

© OECD/IEA, 2013

World

19

0

9

8

1ථ258

969

2ථ642

2ථ524

2 3 4 5 6 7 8 9 10 11 12 13 14 15

In sub-Saharan Africa, the number of people without access to electricity in 2030 is projected to reach 645 million, 8% more than in 2011. It is the only region where the number of people without access to electricity deteriorates over the Outlook period, resulƟng in sub-Saharan Africa’s share of the global total increasing from less than half in 2011 to two-thirds in 2030. Developments in sub-Saharan Africa are not uniform across the 22.ഩ While the Outlook period for tKͲϮϬϭϯ is 2011 to 2035, analysis in this section is based on the period 2011 to 2030, so as to be consistent with the timeframe of the SE4All initiative.

Chapter 2 | Global energy trends to 2035

1

91

16 17 18

projecƟon period, with the rise in the numbers lacking access levelling oī in the 2020s and a decline beginning just before 2030. Brazil is projected to achieve universal access in the next few years – the aim of its ͞Luz para Todos͟ (Light for All) programme (see the Brazil Energy Outlook in PartථB) – while the rest of LaƟn America is projected to achieve universal access around 2020. The number of people relying on the tradiƟonal use of biomass for cooking is projected to drop slightly, to just over 2.5ථbillion in 2030 – around 30% of the global populaƟon at that Ɵme. Economic growth, urbanisaƟon and clean cooking programmes all help improve the situaƟon in developing Asia, where the number of people without clean cooking faciliƟes declines by around 290ථmillion. Despite this, India sƟll has around 730ථmillion people without clean cooking faciliƟes in 2030, equivalent to half of the populaƟon (Figureථ2.17). While the overall picture has improved slightly compared with tKͲϮϬϭϮ, our projecƟons conƟnue to show a worsening situaƟon in sub-Saharan Africa, where nearly 880ථmillion people (63% of the populaƟon), do not have access to clean cooking faciliƟes in 2030. Figure 2.17 ‫ ٲ‬Shares of population with access to electricity and clean cooking facilities by region in the New Policies Scenario Electricity

Clean cooking 2011

Middle East

Increment to 2030 from 2011

Lan America China India Other developing Asia Sub-Saharan Africa 100%

75%

50%

25%

25%

50%

75%

100%

ŶĞƌŐLJĨŽƌůůĂƐĞ

© OECD/IEA, 2013

A trajectory consistent with achieving universal access to electricity and clean cooking faciliƟes by 2030 has been drawn up in the Energy for All Case. To arrive at the required trajectory, in the case of electricity, we assess the required combinaƟon of on-grid, mini-grid (such as village or district level generaƟon)ථand isolated oī-grid soluƟons (such as solarථPV) in each region, taking account of regional costs and consumer density in determining a regional cost per megawaƩ-hourථ(MWh). When delivered through an established grid, the cost per MWh is cheaper than other soluƟons, but extending the grid to remote areas can be very expensive and incur high transmission losses.23 In developing Asia, around 23.ഩ We assume that grid extension is the most suitable option for all urban zones and around 30% of rural areas, but not in more remote rural areas. The remaining rural areas are connected either with mini-grids (65% of this share) or small, stand-alone off-grid solutions (the remaining 35%), which have no transmission and distribution costs. 92

World Energy Outlook 2013 | Global Energy Trends

three-quarters of people gaining access are connected to the main grid or to mini-grid systems. In sub-Saharan Africa, more people gain access through oī-grid soluƟons, as a larger proporƟon of the populaƟon lacking access live in rural areas. In the case of clean cooking faciliƟes, access is also assumed to be achieved through diīerent technologies: one of the most common opƟons is LPG stoves, adopted by 7ථmillion households per year on average in developing Asia and 5ථmillion households per year in sub-Saharan Africa over the projecƟon period.

1

Universal access to modern energy has only a small impact on global energy demand and related CO2 emissions. The addiƟonal energy demand for electricity generaƟon is around 120ථMtoe, pushing total primary energy demand up by less than 1% relaƟve to the New Policies Scenario in 2030͖ but only around 35% of the addiƟonal generaƟon comes from fossil fuels, with the remainder coming from renewables. For cooking, an addiƟonal 0.82ථmbͬd of LPG is required in 2030. The addiƟonal CO2 emissions in the Energy for All Case are negligible, 260ථMt higher in 2030, and only 0.7% higher than in the New Policies Scenario. This small increase in CO2 emissions is aƩributable to the low level of energy per capita expected to be consumed by the people gaining modern energy access and to the relaƟvely high proporƟon of renewable soluƟons adopted. The total impact on greenhouse-gas emissions of achieving universal access to clean cooking faciliƟes needs to be treated with cauƟon, but it is widely accepted that advanced cookstoves, more eĸcient than tradiƟonal biomass stoves, would reduce emissions.

4

Energy subsidies

2 3

5 6 7 8 9 10

ƐƟŵĂƚĞĚĐŽƐƚƐ Subsidies to fossil fuels distort energy markets in many countries, pushing up energy use and emissions, and engendering large economic costsථ(Boxථ2.3). Fossil-fuel consumpƟon subsidies worldwide are esƟmated to have totalled $544ථbillion in 2012. This Įnding is based on a survey that idenƟĮed 40 countries that set energy prices below reference prices, which we deĮne as the full cost of supply based on internaƟonal benchmarks.24 The esƟmates cover subsidies to fossil fuels consumed by end-users and subsidies to fossil-fuel inputs to electric power generaƟon, but do not cover subsidies to petrochemical feedstocks. Unlike oil, gas and coal, electricity is not extensively traded over naƟonal borders, so subsidy esƟmates are based on the diīerence between end-user prices and the cost of electricity producƟon, transmission and distribuƟon. Countries that subsidise fossil fuels fall into two broad groups: those that import energy at world prices and then sell it domesƟcally at lower regulated prices͖ and those that are net exporters of energy — and therefore do not import energy at world prices — but price it domesƟcally at below the reference prices.

11 12 13 14 15 16

© OECD/IEA, 2013

17 24.ഩ Some authorities regard the use of international benchmark prices to calculate reference prices as inappropriate. In particular, some are of the opinion that reference prices should be based on actual production costs, particularly when estimating subsidies in energy resource-rich countries, rather than prices on international markets as applied within this analysis.

Chapter 2 | Global energy trends to 2035

93

18

Box 2.3 ‫ ٲ‬Smuggling as a possible driver of fossil-fuel subsidy reform The prevalence of fossil-fuel subsidies in many parts of the world is making fuel smuggling a serious problem by providing an incenƟve to sell subsidised products from one country in neighbouring countries where prices are higher. While the smugglers make big Įnancial gains, there is a high Įnancial cost to the subsidising country (with no naƟonal beneĮt), and probably substanƟal Įnancial costs to the recipient country, by way of forgone taxes and excise duƟes, due to reduced legiƟmate sales. A good example is Iran. Although subsidy reforms in 2010 reduced the incenƟve to smuggle gasoline and diesel to neighbouring countries, a sharp devaluaƟon of its currency against the US dollar in 2013 (which was not matched by an adjustment in local prices) has increased it again. It is esƟmated that around 10ථmillion litres of fuel, or more than 60 000ථbarrels, are currently being smuggled out of the country each day, mainly into Pakistan. It is potenƟally a very lucraƟve acƟvity, as diesel sƟll sells for as liƩle as $0.12 per litre in Iran, compared with $1.20 per litre in Pakistan. In Southeast Asia, the geographical proximity of countries with major retail price dispariƟes has meant that fuel smuggling has long been a major problem. It oŌen involves the use of small oil tankers or Įshing boats that either bypass normal customs routes altogether or falsely declare their load as products that are exempt from excise duƟes. Gasoline in Indonesia, for example, was unƟl recently around 60% cheaper than in a number of its neighbouring countries. Subsidies in Malaysia have reduced reĮned product prices to well below the regional average. In the Philippines, which has been the recipient of a lot of smuggled fuel, the government esƟmates that its tax revenues are being reduced by around $1ථbillion per year as a result of illegiƟmate purchases.

© OECD/IEA, 2013

Many countries are taking steps to stamp out fuel smuggling. Saudi Arabia, for example, has increased inspecƟons of vehicles leaving its borders to check that they have only enough fuel to get to the nearest re-fuelling staƟon on the other side. The Philippines has also recently stepped up coastal patrols, in this case to stop smuggled fuel from geƫng into the country. But history has shown that eīorts to curtail smuggling absorb scarce administraƟve resources and are rarely completely successful. While beƩer border control may be a necessary opƟon for countries that are the recipients of smuggled fuels, a much more eīecƟve strategy would be for the originaƟng countries to remove the subsidies, as that would eliminate the incenƟve to smuggle the fuels. The value of fossil-fuel subsidies increased in 2012 compared with 2011, as moderately higher internaƟonal prices and increased consumpƟon of subsidised fuels oīset considerable progress in reining in subsidies in some countries. Oil products were the most heavily subsidised fuels in 2012 and cost $277ථbillion, or 51% of the total. Subsidies to natural gas and coal consumed by end-users amounted to $124 billion and $7 billion respecƟvely. Subsidies to electricity stood at $135ථbillion. Almost all consumpƟon subsidies are in non-OECD countries, while producƟon subsidies, typically intended to 94

World Energy Outlook 2013 | Global Energy Trends

expand domesƟc supply, are a much more common form of subsidy in OECD countries than consumpƟon subsidies (OECDͬIEA, 2013) (Figureථ2.18). ConsumpƟon subsidies remain most prevalent in net energy-exporƟng countries: they accounted for around 75% of the global total in 2012. Figure 2.18 ‫ ٲ‬Economic value of fossil-fuel consumption subsidies by fuel for

1 2 3

top 25 countries, 2012

4

Oil

Iran

Natural gas

Saudi Arabia Russia

5

Coal

India

Electricity

Venezuela

6

China Indonesia

7

Egypt UAE

8

Iraq Mexico Algeria

9

Pakistan Uzbekistan

10

Argenna Kuwait Ukraine

11

Thailand Malaysia Qatar

12

Bangladesh Turkmenistan

13

Kazakhstan Ecuador Nigeria

© OECD/IEA, 2013

10

20

30

40

50

60

14

70 80 90 Billion dollars (2012)

Subsidies to renewable energy are aimed chieŇy at improving the compeƟƟveness of renewables ǀŝƐͲăͲǀŝƐ convenƟonal alternaƟves. Our latest esƟmates show that renewables subsidies increased by 11% to reach $101 billion in 2012, primarily due to the increase in solar PV capacity, but that they conƟnue to be less than one-ĮŌh of the level of fossilfuel consumpƟon subsidiesථ(see Chapterථ6). While some renewable energy technologies, such as hydropower and geothermal, have long been economic in many locaƟons, others, such as wind (parƟcularly oīshore) and solar, require Įnancial support to foster their deployment in most countries. SigniĮcant growth in renewables is projected in the New

Chapter 2 | Global energy trends to 2035

95

15 16 17 18

Policies Scenario, mainly driven by subsidies, which are projected to conƟnue to rise to around $220ථbillion in 2035, although they start declining before then in some regions as diīerent technologies become compeƟƟve. Support schemes for renewable energy need to be carefully designed (and someƟmes re-designed) if they are to achieve their objecƟve in the most cost-eīecƟve way.

hƉĚĂƚĞŽŶĨŽƐƐŝůͲĨƵĞůƐƵďƐŝĚLJƌĞĨŽƌŵ

© OECD/IEA, 2013

A number of major reforms to reduce or phase out fossil-fuel subsidies have been announced since last year’s Outlook, signiĮcantly adding to the momentum that has been building up over recent years. Barring a major increase in internaƟonal energy prices or in consumpƟon, these reforms – if they prove durable – will lead to a reducƟon in the economic cost of fossil-fuel subsidies and the associated environmental damage. Economic factors have become the dominant driver of moves to reform fossil-fuel subsidies as rising consumpƟon and persistently high energy prices have made them an unsustainable Įnancial burden in many instancesථ(Figureථ2.19). For example, Indonesia increased the prices of gasoline by 44% and diesel by 22% in June 2013 in order to reduce the strain on the state budget. The last Ɵme fuel prices were raised in Indonesia was in 2009 and since then the cost of subsidies has risen in line with the country’s mounƟng dependence on imported oil and a boom in vehicle ownership in the fast-growing economy. The reforms, which were accompanied by cash hand-outs to poor households, have proved successful. Although providing blanket subsidies to an enƟre populaƟon is an extremely ineĸcient way to make energy aīordable for the poor, if the subsidies are to be removed, it is oŌen important to provide targeted welfare assistance to avoid restricƟng access to modern energy services. Other parƟcularly notable reforms to energy pricing were made by India and China during the year. India has started increasing diesel prices on a monthly basis (with the eventual goal of eliminaƟng subsidies enƟrely) following reforms to gasoline pricing that were introduced in 2010. India has also announced that power staƟons that need to buy imported coal due to local supply shorƞalls will be able to pass on the extra costs to their customers. Under the old system, tariīs could not be increased to reŇect fuel prices, someƟmes leaving generators with liƩle incenƟve to increase generaƟon to meet peak demand and so contribuƟng to frequent blackouts and rolling outages. India has also announced that prices of domesƟcally produced natural gas will be adjusted on a quarterly basis from April 2014, to match the average of the prices of the LNG it imports and of gas on other major internaƟonal markets. This is expected to result in a doubling of domesƟc gas prices. In a similar move, China increased natural gas prices by 15% for non-residenƟal users, which make-up around 80% of total demand. In both countries, the reforms increase the incenƟve to produce gas domesƟcally and make it more economic to import gas, helping to meet targets to increase the share of gas in the energy mix. In Russia, a shiŌ towards market-based pricing in recent years has seen a narrowing of the gap between domesƟc gas and electricity prices and comparable internaƟonal levels.

96

World Energy Outlook 2013 | Global Energy Trends

© OECD/IEA, 2013

Chapter 2 | Global energy trends to 2035

Figure 2.19ථ‫ ٲ‬Rates of fossil-fuel consumption subsidies in 2012 and recent developments in selected countries China

Russia 3 Raised electricity and gas prices by 15% on average in July 2013, and plans to increase them further in July 2015.

3 Increased natural gas prices for non-residential users by 15% in July 2013. 3 Implemented a tiered electricity pricing system in July 2012.

Iran

Jordan

3 tarted a programme in late 2010 to gradually phase out subsidies by 2015. After delays to the second-phase, plans to further increase fuel prices in 2013.

3 Increased electricity tariffs by up to 15% for some sectors in August 2013, and plans to increase tariffs for households from 2014.

Egypt 3 Increased prices of fuel oil, gas and diesel for industry in February 2013. 3 Plans to implement a smart card system to track subsidised fuel in the beginning of 2014.

Thailand 3 rom )-ember 2013, increasing LPG prices monthly. 3 Increased electricity tariffs in )-ember 2013, and will revise them every four months.

Malaysia 3 n )-ember 2013, raised gasoline and diesel prices. 3 Plans to implement in 2014 a subsidy removal programme set out in 2011 to increase natural gas and electricity prices.

Mexico 3 Has been raising the prices of gasoline and diesel monthly in 2013 to bring them closer to international levels.

India Rate of subsidisation Above 50%

South Africa Between 20% and 50%

3 n February 2013, started implementing an 8% annual average electricity price increase over the next five years.

Below 20%

3 Plans to nearly double natural gas prices from April 2014, and to revise them quarterly until 2017. 3 Has been raising the diesel price each month in 2013. 3 ower companies allowed to increase electricity prices from June 2013 to reflect the costs of imported coal.

Indonesia 3 Has raised electricity tariffs each quarter in 2013. 3 n June 2013, increased the price of gasoline and diesel by 44% and 22% respectively.

97

This map is without prejudice to the status of or sovereignty over any territory, to the delimitation of international frontiers and boundaries and to the name of any territory, city or area.

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

© OECD/IEA, 2013

MulƟlateral co-operaƟon to support fossil-fuel subsidy reform has also conƟnued to build throughout 2013. Four years have now passed since leaders of the G-20 and AsiaPaciĮc Economic CooperaƟon forum (APEC) commiƩed to ͞phase out ineĸcient fossil-fuel subsidies that encourage wasteful consumpƟon͟.25 Much remains to be done to fulĮl these commitments, but many countries introduced reforms aimed at reducing subsidies, while G-20 countries have begun undertaking voluntary peer reviews of each other’s subsidies and reform eīorts. In a more recent iniƟaƟve, a high-level panel reporƟng to the United NaƟons Secretary General has recommended that a goal of achieving sustainable energy be included in the post-2015 Development Agenda, which will succeed the Millennium Development Goals, and that this objecƟve should encompass the phasing out of fossilfuel subsidies.

25.ഩ Since the G-20 Pittsburgh Summit in 2009 committed to ͞rationalize and phase out over the medium term inefficient fossil fuel subsidies that encourage wasteful consumption͟, the IEA along with several other international organisations has been providing the group with ongoing analysis aimed at supporting the implementation of their commitment. 98

World Energy Outlook 2013 | Global Energy Trends

Chapter 3 Natural gas market outlook What price is right? Highlights

x Although current market condiƟons vary markedly across the world, the overall outlook for natural gas is bright: consumpƟon in 2035 is higher than in 2012 in all three scenarios. In the New Policies Scenario, gas use rises by 1.6% per year on average to reach almost 5ථtcm in 2035͖ 82% of this increase is concentrated in non-OECD countries, where demand rises by around 1.3ථtcm.

x The biggest absolute increases in demand are in China, the Middle East (where gas use overtakes that of the European Union before 2020) and North America. Outside Japan,
x New sources of gas, both convenƟonal and unconvenƟonal, bring addiƟonal diversity to global supply over the Outlook period. New contributors to convenƟonal output growth include Iraq, East Africa, Brazil and the eastern Mediterranean, supplemenƟng increases from established suppliers in Russia, the Caspian, North and West Africa and the Middle East. UnconvenƟonal gas accounts for almost half of the growth in global output and its development spreads well beyond North America, notably aŌer 2020, making China and Australia major contributors to global producƟon growth.

x Changes in the cast of major LNG suppliers create new linkages between regional gas markets, notably between those of North America and the Asia-PaciĮc, narrowing to a degree the wide regional gas price diīerenƟals that exist today. LNG exports from the United States factor strongly in puƫng addiƟonal pressure on tradiƟonal oil-indexed mechanisms for pricing gas and in loosening the current rigidity of LNG contracƟng structures, although various market and insƟtuƟonal barriers conƟnue to put a brake on global gas market integraƟon.

© OECD/IEA, 2013

x We examine a Gas Price Convergence Case in which convergence between diīerent regional gas prices is more rapid than in the New Policies Scenario, underpinned by the assumpƟon of higher LNG volumes from North America, faster changes to gas markets and pricing mechanisms in the Asia-PaciĮc region, and an easing of costs for liquefacƟon and LNG shipping. In this case, lower prices result in higher global gas demand (by 107ථbcm in 2035) and reduced import bills. Prices remain suĸciently aƩracƟve to bring forward addiƟonal producƟon from a range of suppliers.

Chapter 3 | Natural gas market outlook

99

Global overview Whatever the policy landscape for the next quarter of a century, natural gas is set to grow in importance globally thanks to its widespread availability, compeƟƟve supply costs and environmental advantages over the other fossil fuels. Since the turn of the century, global gas use has expanded on average by 2.7% per year – faster than oil and nuclear power, but more slowly than coal and renewables. The share of gas in the world energy mix conƟnues to rise, with unconvenƟonal gas playing an increasingly signiĮcant role in meeƟng growing gas demand. zet behind this upbeat global outlook for gas, there are marked variaƟons by region – with gas use in Europe, in parƟcular, facing a more diĸcult future. These regional dispariƟes are caused by diīerences in the dynamics of inter-fuel compeƟƟon and speciĮc economic and policy condiƟons. Table 3.1 ‫ ٲ‬Natural gas demand and production by region and scenario (bcm) ථ





New Policies

Current Policies

450 Scenario

1990

2011

2020

2035

2020

2035

2020

2035

1ථ036

1ථ597

1ථ707

1ථ885

1ථ741

1ථ999

1ථ654

1ථ493

881

1ථ195

1ථ358

1ථ483

1ථ377

1ථ585

1ථ334

1ථ237

Demand

1ථ003

1ථ773

2ථ249

3ථ086

2ථ291

3ථ279

2ථ149

2ථ554

ProducƟon

1ථ178

2ථ188

2ථ599

3ථ492

2ථ655

3ථ693

2ථ472

2ථ817

World*

Demand

2ථ039

3ථ370

3ථ957

4ථ976

4ථ032

5ථ278

3ථ806

4ථ054

Share of

Demand

49%

53%

57%

62%

57%

62%

56%

63%

non-OECD

WƌŽĚƵĐƟŽŶ

57%

65%

66%

70%

66%

70%

65%

69%

ථ OECD Non-OECD

Demand ProducƟon

* For 1990 and 2011, the world numbers shown correspond to demand. For the projecƟons, demand and producƟon are always the same, as stock changes are assumed to be zero. The world numbers include gas use as an internaƟonal marine fuel. Note: bcm с billion cubic metres.

© OECD/IEA, 2013

In the New Policies Scenario, the share of gas in the global energy mix reaches 24% in 2035, up from 21% in 2011 (almost catching up with coal in the process), but the pace of annual gas demand growth, which averages 1.6% per year, declines progressively through the projecƟon period. In the Current Policies Scenario, demand grows faster – at 1.9% per year – as no new policies are introduced to rein in demand for either gas or electricity, resulƟng in stronger demand for gas to generate power. In the 450 Scenario, demand grows by only 0.8% per year, levelling oī in the late 2020s, with consumpƟon in the power sector especially subdued. Regardless of the scenario, future gas demand growth is led by nonOECD countries. Their share of global demand already overtook that of the OECD in 2007 and reaches 62% in 2035 (up from 53% in 2011) (Tableථ3.1). Non-OECD countries also account for the bulk of the growth in gas producƟon across the three scenarios. In the OECD, producƟon in North America and Australia grows briskly, with both regions becoming major gas exporters͖ but producƟon falls in Europe. UnconvenƟonal gas grows strongly in all scenarios, accounƟng for 27% of total producƟon in 2035, compared with 17% in 2011. New sources of convenƟonal producƟon also emerge, notably in Iraq, 100

World Energy Outlook 2013 | Global Energy Trends

East Africa and the deepwater eastern Mediterranean. This diversity of supply sources and of supply routes can be expected to contribute to an environment of growing conĮdence in the adequacy and reliability of gas supply. Box 3.1 ‫ ٲ‬Wide variety in regional starting points for the gas outlook The last year has seen extreme divergence in market condiƟons across the major regional markets (trends since 2005 for selected countries and regions are shown in Figureථ3.1). The gas market in North America remains characterised by ample supply of natural gas and low prices, which have permiƩed gas to gain market share in the power sector at the expense of coal. Despite much higher prices, there has also been strong growth in gas consumpƟon across much of Asia. In China, which in 2011 became the third-largest individual gas market in the world aŌer the United States and Russia, demand has been driven primarily by policy intervenƟons, while gas demand in Japan has been boosted by the need to replace lost power generaƟon due to the shutdown of the nuclear Ňeet following the events at Fukushima Daiichi. In the Middle East, rapid consumpƟon growth has been sƟmulated in many instances by low prices that do not reŇect the internaƟonal value of the gas. ProducƟon has also risen substanƟally in the Middle East region as whole, but, outside Qatar, it has oŌen struggled to keep up with demand. By contrast, condiƟons in Europe have remained diĸcult with gas use declining by a further 2% in 2012, as economic condiƟons depressed overall energy demand, and increased renewables supply and cheaper coal-Įred generaƟon (aided by depressed CO2 prices) backed out gas in the power sector. Figure 3.1ථ‫ ٲ‬Natural gas demand and production growth in selected

3 4 5 6 7 8 9 10

12

United States Producon

2

11

regions, 2005-2012

Rest of Middle East

13

Qatar China Russia

14

Middle East

15

United States Demand

1

China

16

Japan and Korea European Union

© OECD/IEA, 2013

-60

-30

0

30

60

90

120

150

17

180 bcm

18 Chapter 3 | Natural gas market outlook

101

Demand Regional trends In the New Policies Scenario (on which this analysis concentrates), the fastest-growing gas markets in the world are all outside the OECD. Non-OECD countries account for more than three-quarters of global primary demand growth over the period to 2035, with the biggest increases in absolute terms occurring in China and the Middle East. ConsumpƟon increases but rates of growth are smaller in the three main OECD regions, because of saturaƟon eīects and strong penetraƟon of renewables in the power sector in Europe. Nonetheless, these OECD markets remain comparaƟvely large. For example, demand in the United States, which remains the world’s single largest gas consuming country, is 50% higher than Chinese demand in 2035. Figure 3.2 ‫ ٲ‬Natural gas demand in selected regions in the New Policies Scenario 2035

United States

2011

Middle East European Union Russia China India Japan

© OECD/IEA, 2013

100

200

300

400

500

600

700

800 bcm

Despite relaƟvely low gas prices, the maturity of the United States and Canada as gas markets limits the scope for rapid growth in North American natural gas demand, even though price diīerenƟals with other fuels do create incenƟves to expand gas use into new areas, such as transport. For the region as a whole (including Mexico, where growth is faster), gas demand rises from 864ථbillion cubic metre (bcm) in 2011 to 1ථ036ථbcm in 2035. The main driver for this increase is electricity generaƟon. Whereas the short-term rise in gas use is largely due to switching from coal to gas in exisƟng plants (for which 2012 may have represented a high-water mark), the longer-term trend depends more on what type of new capacity is built: we see gas as the preferred fuel for new thermal generaƟon, as environmental restricƟons limit the scope for building new coal plants. Outside the power sector, the transport sector sees a rapid rate of increase in demand, though it sƟll accounts for only a relaƟvely small share of total gas use in 2035, as it starts from a very low base. (The prospects for natural gas use in transport are discussed in Chapter 15).

102

World Energy Outlook 2013 | Global Energy Trends

Table 3.2 ‫ ٲ‬Natural gas demand by region in the New Policies Scenario (bcm)

OECD Americas United States

1990

2011

2020

2025

2030

2035

1 036

1 597

1 707

1 778

1 827

628

869

957

988

533

696

749

769

2011-2035 Delta

CAAGR*

1 885

289

0.7%

1 016

1 044

175

0.8%

781

789

93

0.5%

325

525

537

568

584

605

80

0.6%

82

202

214

222

227

236

34

0.6%

57

120

119

123

122

124

3

0.1%

1 003

1 773

2 249

2 541

2 815

3 086

1 313

2.3%

738

703

732

756

785

817

114

0.6%

Caspian

100

117

127

134

139

144

27

0.9%

Russia

447

476

493

504

523

544

68

0.6%

Europe Asia Oceania Japan Non-OECD E. EuropeͬEurasia

84

410

669

816

949

1 088

678

4.2%

China

15

132

307

396

470

529

397

6.0%

India

13

61

87

114

140

172

111

4.4%

Asia

Middle East

87

399

504

577

645

700

301

2.4%

Africa

35

111

153

170

187

204

93

2.6%

LaƟn America

60

149

190

221

248

277

128

2.6%

4

27

45

61

75

90

63

5.2%

2ථ039

3ථ370

3ථ957

4 322

4 646

4 976

1 606

1.6%

371

492

494

523

537

554

62

0.5%

Brazil World** European Union

* Compound average annual growth rate. ** The world numbers include gas use as an internaƟonal marine fuel.

© OECD/IEA, 2013

The outlook for gas demand in Europe remains subdued. Demand in OECD Europe fell to 514 bcm in 2012 – the second consecuƟve year of decline and down 10% from 2010 – and is now back to the level of 2003. The situaƟon is similar within the European Union. The weak economic environment and high gas prices are the main causes, but a combinaƟon of low coal prices, rock-boƩom prices for carbon-dioxide (CO2) and the big expansion in renewables-based capacity, plus eĸciency and energy saving measures, have also played their part. In the New Policies Scenario, demand in OECD Europe recovers only very slowly, returning to 2010 levels only around 2025 before reaching just over 600ථbcm in 2035. The power sector holds the key to gas demand in Europe and the prospects in this area depend on the relaƟonship between gas, coal and carbon prices (see Chapterථ5, Box 5.1 for analysis of coal-to-gas switching in the power sector). A gradual re-balancing in relaƟve prices favours gas, notably because the extremely low carbon prices in Europe seen in recent years are not expected to persist: we assume that they will rise from the current level of around $6 per tonne (in mid-2013) to $20ͬtonne by 2020 and $40ͬtonne by 2035. In addiƟon, a number of coal plants are expected to close in Europe as a result of new air-quality legislaƟon at European level. Europe’s nuclear capacity is also expected to tail oī as, in aggregate, more plants are de-commissioned than built. These factors help to Chapter 3 | Natural gas market outlook

103

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18

keep gas use in absolute terms on a rising trend in the New Policies Scenario. But there are sƟll signiĮcant quesƟons about European gas demand prospects, not least of which is the uncertainty over the business case for new convenƟonal capacity in a power market characterised by a rising share of renewables (see Chapterථ6). The European energy policy targets for 2030 could also herald a new eīort on energy eĸciency and CO2 emissions reducƟons, both of which could squeeze gas demand further.1 In Japan, energy policies are a key uncertainty for the outlook for gas demand. In parƟcular, the future role of nuclear power remains unclear at the Ɵme of wriƟng. The projecƟons of the New Policies Scenario assume greater emphasis in Japan on energy eĸciency and renewables, but with gas nonetheless conƟnuing to play an important role, parƟcularly in meeƟng peak demand. Gas demand is projected to stabilise at around 120-125ථbcm per year. In the short term, unavoidable reliance on expensive liqueĮed natural gas (LNG) purchases to compensate for the shorƞall in nuclear output has added to Japanese import bills, strengthening the dissaƟsfacƟon within Japan about the high oil-indexed prices that currently form the basis for gas trade in the Asia-PaciĮc region. Russia, the world’s second-largest gas consumer, faces considerable uncertainty over domesƟc demand, relaƟng primarily to the rate and direcƟon of price reform and the speed at which it’s ageing, highly ineĸcient energy-consuming capital stock will be replaced. The potenƟal for eĸciency gains is enormous and this is realised, in part, in the New Policies Scenario: Russian demand grows only slowly, from 476ථbcm in 2011 to around 545 bcm by 2035, with gas demand in the power sector Ňat (at around 285ථbcm), even though generaƟon from gas-Įred power plants increases by one-quarter (140ථterawaƩ-hours ΀TWh΁). The modest growth in other sectors is driven by an expanding economy and the expectaƟon of expanded average residenƟal space per capita. A more rapid shiŌ towards market-based pricing for industrial and residenƟal users and greater aƩenƟon to energy eĸciency policies could easily bring Russian gas demand to a Ňat or declining trajectory. Price changes are also expected to curb demand in some neighbouring markets. In Ukraine, for example, higher prices for imported gas have dampened domesƟc demand (which fell by 8% in 2012) and created stronger incenƟves to develop the country’s indigenous gas resources, including unconvenƟonal resources.

© OECD/IEA, 2013

In the New Policies Scenario, China sees by far the largest increase in gas demand in any single country, reaching 530ථbcm in 2035. As in almost all other regions, electricity generaƟon is the main source of addiƟonal demand: the widely shared concerns about air quality and local pollutants among China’s rapidly expanding urban populaƟon make a forceful case for gas, rather than coal, as the preferred fuel for powering the country’s ciƟes. More than 60ථgigawaƩs (GW) of gas-Įred capacity is due to be online by the end of the current Įveyear plan in 2015͖ around 50ථbcm of gas demand can be expected from this source alone. 1.ഩ The role of gas as back-up for variable renewable power sources implies a significant volume of gas-fired capacity to be available, but it does not require this capacity to operate for many hours in the year. Thus, in the 450ථScenario, the European Union requires 280ථGW of gas-fired power in 2035, but this generates only 370ථTWh at 15% capacity utilisation, compared with 800ථTWh at almost 30% utilisation in the New Policies Scenario͖ the difference in gas demand between the two scenarios is around 75ථbcm. 104

World Energy Outlook 2013 | Global Energy Trends

Over the projecƟon period as a whole, gas use in the power sector is projected to increase six-fold, to around 160ථbcm, driven by environmental policies (including the introducƟon of carbon pricing from 2020). Road transport represents another area of nascent, but potenƟally rapid growth͖ consumpƟon of around 10ථbcm in 2011 is set to triple over the projecƟon period, driven by air quality and energy security concerns. In India, gas consumpƟon is expected to remain constrained in the short to medium term by low availability of domesƟc gas producƟon and the high cost of imported LNG. However, consumpƟon picks up again in the laƩer part of the decade, as the supply situaƟon improves, with the power sector leading the way and accounƟng for almost half of total gas use by 2035 (80ථbcm out of a total of 170ථbcm). ConsumpƟon in the transport sector also increases strongly, to reach 18ථbcm: India is already one of the global pace-seƩers for natural gas vehicles. Outside Asia, the Middle East sees the biggest increase in gas demand in absolute terms – around 300ථbcm – between 2011 and 2035, driven by new power generaƟon, water desalinaƟon and petrochemical projects. Gas has become a popular fuel across the region, parƟcularly because it has usually been available at low cost as a by-product of oil producƟon (and because it provides an alternaƟve to oil consumpƟon, freeing up the more valuable product for export). However, this has led to imbalances in some markets, with gas output lagging behind fast-growing demand, sƟmulated by low domesƟc gas prices which do not reŇect the internaƟonal value of the gas.
© OECD/IEA, 2013

2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Sectoral trends The power sector remains the main driver of increased gas demand worldwide in the New Policies Scenario, though consumpƟon trends are highly sensiƟve to compeƟƟve pressures from other fuels, notably coal and renewables, and to the impact of government policies. Outside major gas-producing countries and regions (including North America), coal is generally less expensive than gas as a fuel for generaƟng electricity (especially in Chapter 3 | Natural gas market outlook

1

105

17 18

the absence of a price for CO2) in both exisƟng and new plants. But gas use for power nonetheless conƟnues to expand, albeit at varying rates by region, as it has a number of advantages that make it aƩracƟve to investors and policymakers alike. These include high technical eĸciency and Ňexibility (making it suitable for complemenƟng variable renewables), the relaƟve ease and speed of construcƟon and its low carbon and other emissions characterisƟcs, compared with coal and oil. Moreover, the up-front capital expenditures tend to be lower for gas than coal plants. In the New Policies Scenario, gas use for power grows by around 42% and electricity generaƟon remains the leading gasconsuming sector (Figureථ3.3). The increase is especially marked in the Middle East (where it doubles) and China (where it expands six-fold), and India (where it more than triples). Figure 3.3 ‫ ٲ‬World natural gas demand by sector in the New Policies Scenario Final sectors

bcm

Transformaon sectors 2 500 2 000

Increment 2020-2035 1.5%

Increment 2011-2020 2011

1 500 1.9%

1 000

% CAAGR* 2011-2035

1.3%

1.8% 500

2.9% Power Other Industry generaon transformaon

Buildings

Transport

2.0%

Non-energy use

© OECD/IEA, 2013

* Compound average annual growth rate.

Globally, gas use in the energy sector itself, mainly for oil and gas extracƟon, LNG liquefacƟon and conversion from gas-to-liquids (GTL), together with its use as chemical feedstock expands by more than half, or around 300ථbcm. Most of the increase in feedstock use is directed towards the producƟon of ammonia, the most important base product for ferƟlisers, and of methanol, which is used to produce a variety of products, mainly in the chemicals industry. Outside the power and energy sectors, gas use in industry grows the most in volume terms over the period to 2035 (by 340ථbcm), with most of the increase occurring before 2025. One-third of the growth to 2035 comes from China, with demand growing parƟcularly rapidly (by 14% per year on average) to 2020. Gas demand in the buildings sector (residenƟal and commercial use) grows in OECD countries (by 0.7% per year), but saturaƟon eīects limit the potenƟal for more gas use͖ the increase comes as most of the remaining demand for oil is squeezed out of this sector and space heaƟng demand increases modestly. In non-OECD countries, gas consumpƟon in the buildings sector grows by 75%. In this sector, too, it is China that dominates the picture, accounƟng for almost half of the total non-OECD increase͖ urbanisaƟon and rising incomes, which boost demand for water heaƟng, cooking and space heaƟng, are the main drivers.

106

World Energy Outlook 2013 | Global Energy Trends

The fastest projected rate of growth in gas use (though the smallest sectoral increase in absolute terms) is in the transport sector, with most of the increase coming from road vehicles (see also Chapter 15). The technologies are well-proven and the number of natural gas vehicles (NGVs) increased from about 1.3ථmillion in 2000 to an esƟmated 13.7 million in 2012. But these numbers pale in comparison with the number of vehicles that run on liquid fuels – more than 1 billion. Two-thirds of NGVs on the road today are in non-OECD countries, mostly in Asia and LaƟn America͖ within the OECD, only Italy and
2 3 4 5 6 7 8 9 10

ProducƟon Resources and reserves

11

The world’s remaining resources of natural gas are more than suĸcient to meet any conceivable level of gas demand for the next several decades. Proven reserves of gas stood at 187ථtrillion cubic metres (tcm) at the end of 2012 (BP, 2013). This is marginally lower than the esƟmate one year earlier, as producƟon in 2012 outstripped reserve addiƟons from new discoveries and reassessments of reserves in previously discovered Įelds. Reserves increased sharply in 2011.

© OECD/IEA, 2013

1

Proven reserves are a very narrow indicator of the size of the resource base. Our modelling of gas producƟon is based primarily on esƟmates of technically recoverable resources, which are much larger and which have been increasing over Ɵme.2 At the end of 2012, total remaining technically recoverable resources of gas stood at 810ථtcm, i.e. more than four Ɵmes larger than proven reserves and equivalent to around 235 years of producƟon at current rates (Tableථ3.3). Our latest Įgures take into account the new global assessment of shale gas resources from the US Energy InformaƟon AdministraƟon (US EIA, 2013), which shows an increase of nearly 10% over the esƟmate in their 2011 report, mainly because the latest study covers more geological formaƟons in a larger number of countries. CumulaƟve

12 13 14 15 16 17 18

2.ഩ See Chapter 13 for definitions of the different categories of hydrocarbon resources.

Chapter 3 | Natural gas market outlook

107

gas producƟon to date amounts to some 109ථtcm, meaning that around 12% of ulƟmately recoverable resources have been produced. In the New Policies Scenario, an addiƟonal 100ථtcm is projected to be produced, implying that more than three-quarters of ulƟmately recoverable resources would sƟll remain to be recovered as of 2035. In pracƟce, further upward revisions to resource esƟmates are likely as our understanding of the resource base – notably for unconvenƟonal gas – improves. Table 3.3 ‫ ٲ‬Remaining technically recoverable natural gas resources by type and region, end-2012 (tcm) ConvenƟonal

Total

UnconvenƟonal Tight gas

Shale gas

Coalbed methane

Sub-total

E. EuropeͬEurasia

143

11

15

20

46

190

Middle East

124

9

4

-

13

137

Asia-PaciĮc

44

21

53

21

95

138

OECD Americas

46

11

48

7

66

112

Africa

52

10

39

0

49

101

LaƟn America

32

15

40

-

55

86

OECD Europe

26

4

13

2

19

46

468

81

212

50

343

810

World

Notes: Remaining resources comprise known reserves, reserves growth and undiscovered resources. UnconvenƟonal gas resources in regions that are richly endowed with convenƟonal gas, such as Eurasia or the Middle East, are oŌen poorly known and could be much larger. Sources: BGR (2012)͖ US EIA (2013)͖ USGS (2000)͖ USGS (2012a and 2012b)͖ IEA databases and analysis.

WƌŽĚƵĐƟŽŶƚƌĞŶĚƐ

© OECD/IEA, 2013

In the New Policies Scenario, natural gas producƟon increases in every region of the world between 2011 and 2035 with the excepƟon of Europe, where robust producƟon from Norway is not enough to oīset the decline of maturing Įelds in other parts of the North Sea and onshore Netherlands. ConvenƟonal gas as a whole contributes 52% of the increase in supply, with the rest coming from unconvenƟonal sources (covered in more detail in the next secƟon). China, the United States, Russia and Australia are the countries with the biggest increases in gas output (Tableථ3.4). In North America, rising producƟon of unconvenƟonal gas more than oīsets a decline in convenƟonal gas output and its share of the region’s gas producƟon increases to 70% by 2035. Total gas output in the United States increases by around 190ථbcm, reaching nearly 840ථbcm in 2035, the country remaining the top gas producer globally throughout the projecƟon period. Canadian producƟon is also expected to grow, though more slowly than in the United States, with unconvenƟonal gas similarly oīseƫng a decline in convenƟonal gas output. Mexican producƟon reaches 80ථbcm, with both convenƟonal and unconvenƟonal gas contribuƟng to a 30ථbcm increase in output.

108

World Energy Outlook 2013 | Global Energy Trends

Table 3.4 ‫ ٲ‬Natural gas production by region in the New Policies Scenario (bcm) 2011-2035

1990

2011

2020

2025

2030

2035

Delta

CAAGR*

OECD

881

1ථ195

1ථ358

1ථ403

1ථ430

1ථ483

288

0.9%

Americas

643

859

1ථ000

1ථ041

1ථ063

1ථ114

255

1.1%

Canada

109

160

184

189

186

194

34

0.8% 2.1%

Mexico United States Europe Norway Asia Oceania

26

49

50

58

68

81

32

507

649

764

792

807

837

188

1.1%

211

277

249

237

225

215

-62

-1.1%

28

101

121

118

115

111

10

0.4%

28

59

109

125

143

155

95

4.1%

Australia

20

51

103

120

139

152

101

4.6%

Non-OECD

1ථ178

2ථ188

2ථ599

2ථ919

3ථ216

3ථ492

1ථ304

2.0%

831

882

911

986

1ථ094

1ථ164

282

1.2%

10

16

23

33

43

47

30

4.5%

629

673

667

692

757

808

135

0.8%

85

67

83

100

117

132

66

2.9%

130

419

566

625

694

769

350

2.6%

China

15

103

178

218

266

317

214

4.8%

India

13

46

62

73

85

98

52

3.2%

Indonesia

48

81

108

118

129

139

57

2.3%

Middle East

E. EuropeͬEurasia Azerbaijan Russia Turkmenistan Asia

92

519

624

720

766

823

304

1.9%

Iran

23

150

143

165

180

207

56

1.3%

Iraq

4

6

39

71

79

83

77

11.5%

Qatar

6

151

187

214

227

237

86

1.9%

Saudi Arabia

26

86

112

121

128

136

50

1.9%

UAE

20

52

58

61

62

65

13

0.9%

Africa

64

200

280

333

378

428

228

3.2%

43

77

106

115

123

132

55

2.3%

6

8

17

21

24

30

22

5.7%

Algeria Libya

4

36

42

55

70

83

47

3.6%

LaƟn America

Nigeria

60

168

218

255

285

308

140

2.6%

ArgenƟna

20

42

49

65

80

91

49

3.3%

4

17

38

60

78

92

76

7.4%

22

25

36

43

52

63

38

3.9%

2ථ059

3ථ384

3ථ957

4ථ322

4ථ646

4ථ976

1ථ592

1.6%

213

185

135

122

114

104

-80

-2.3%

Brazil Venezuela World European Union

© OECD/IEA, 2013

*ථCompound average annual growth rate.

Norway remains the largest gas producer in Europe: with a porƞolio of large projects in the ArcƟc Barents Sea, it is able to sustain its producƟon at current levels throughout the projecƟon period. Elsewhere in Europe, though, the outlook for convenƟonal gas producƟon

Chapter 3 | Natural gas market outlook

109

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18

is one of conƟnued decline, miƟgated, in part only, by a modest rise in unconvenƟonal output. The decline in United
2011-2020

United States

2020-2035

Russia Australia Qatar Iraq Brazil Turkmenistan Iran Algeria -30

0

30

60

90

120

150

180

210

240 bcm

© OECD/IEA, 2013

Australia’s gas producƟon growth is linked directly to plans to expand exports, which, if realised in full, would see the naƟon rival Qatar as the world’s largest LNG exporter by 2020. More than two-thirds of current global investment in LNG is in Australia, where there are already three LNG export projects operaƟng and a further seven under construcƟon. Of the faciliƟes under construcƟon, three in Queensland are based on coalbed methane (the Įrst LNG projects of this kind), while one uses ŇoaƟng LNG technology.3 However, cost increases have been announced in several of the projects that are underway, with the biggest overruns occurring in the Gorgon and the Australia PaciĮc projects (the unprecedented appreciaƟon of the Australian dollar has been a major contribuƟng factor). Such increases threaten to hold back plans for addiƟonal export projects – especially as there are large investment needs elsewhere in the mining and energy sector. Commitments to new resource developments in Australia have slowed markedly over the last year or so, and the prospects for another round of major Australian LNG projects will depend heavily on how costs evolve, on the deployment of new, potenƟally less costly technologies, such as ŇoaƟng LNG, and on compeƟƟon from other regions, notably North America. We project producƟon to rise to 150ථbcm by 2035, at a slightly slower pace than in last year’s Outlook.

3.ഩ Floating LNG technology uses a purpose-built barge to produce LNG from offshore gas. There would otherwise be technical problems or high capital costs to bring gas to land for liquefaction. 110

World Energy Outlook 2013 | Global Energy Trends

In the Caspian region, the biggest increase in gas output is projected to come from Turkmenistan, where producƟon doubles by 2035, reaching 130ථbcm. Exports to China from Turkmenistan increased to 20ථbcm in 2012 and the progressive rise in capacity of the Central-Asia China pipeline, to a planned 65ථbcm per year, is the main spur for output growth. The Galkynysh Įeld – the second-largest gas Įeld in the world, which began producƟon in mid-2013 – is the main source of incremental output. The increased capacity in the export pipeline from Turkmenistan to China also sƟmulates upstream developments that can tap in along the pipeline route in
© OECD/IEA, 2013

The producƟon outlook in China, which sees output triple to almost 320ථbcm from 103ථbcm in 2011, will depend on progress with unconvenƟonal gas and also on the widespread and Ɵmely implementaƟon of reforms in the pricing of wholesale gas, announced by the Chinese government late in 2011. The reforms are designed primarily to encourage upstream investment, but also to sƟmulate import infrastructure development. While trials have been implemented in two regions, extension of the scheme naƟonally is proceeding only gradually. ASEAN as a whole is poised for a signiĮcant expansion in its gas producƟon, drawing on a large resource base and growing demand (and high prices) for LNG in the Asia-PaciĮc

Chapter 3 | Natural gas market outlook

111

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18

market.4 Output rises from 203ථbcm in 2011 to 265ථbcm in 2035. Indonesia, the largest producer, sees output rise to 140ථbcm, from 81ථbcm in 2011. In the absence of extensive inter-regional pipeline networks, gas resources located far from naƟonal demand centres are likely to be developed as LNG export projects, while those located nearby go to meet domesƟc demand. However, many individual countries within the region will have to turn increasingly to energy imports to saƟsfy domesƟc demand growth, incurring higher energy import bills. Major gas producers, such as Indonesia and Malaysia, will face diĸculty in allocaƟng supply between domesƟc demand and exports that provide an important source of government revenue. Box 3.2 ‫ ٲ‬Natural gas liquids and upstream gas investment5 NGLs are liquids produced within a natural gas stream, separated from the gas Ňow either at the well site (Įeld condensate) or at gas processing plants5. The boost that they provide to upstream gas economics is not a new phenomenon. Gas producƟon in Qatar, for example, has long been driven by the condensate and liquids output, which essenƟally covers all upstream costs. It is also well-documented how producers in the United States have been targeƟng wet-gas plays, where the value of the liquids provides them with economic returns even at very low gas prices. This has allowed natural gas prices to remain lower for longer than many had assumed, an eīect that is likely to conƟnue, at least unƟl the most liquids-rich gas plays start to deplete.

© OECD/IEA, 2013

Russia has long been an outlier in global NGLs producƟon, with a relaƟvely small content of NGLs in gas producƟon to date. The reason for this is that most of the gas produced from the most accessible, uppermost layers of the huge western Siberian Įelds, the tradiƟonal mainstays of Russian producƟon, has been very dry. But this is changing. The aƩracƟon of NGLs among Russian gas players is ampliĮed by relaƟvely favourable tax treatment and by the opportuniƟes for export (in contrast to natural gas, where Gazprom retains, for the moment, an export monopoly). Novatek has been among the leading companies in Russia to invest in gas processing faciliƟes and to target NGLs͖ these liquids accounted for less than 10% of producƟon in energy terms in 2012, but around 30% of company revenues. Gazprom has also been increasing its interest in NGL producƟon, helping to oīset the drop in demand (and price concessions) for its gas in Europe by producing from deeper, weƩer, layers at its mainstay gas Įelds in western Siberia. Over the projecƟon period, we anƟcipate that Russia sees a gradual growth in the share of NGLs in its produced gas, moving it into line with the situaƟon in many other countries where NGLs make a major contribuƟon to the economics of upstream gas projects.

4.ഩ The prospects for ASEAN energy markets are discussed in detail in Southeast Asia Energy Outlook: World Energy Outlook Special Report released in October 2013 (IEA, 2013a). 5.ഩIn our projections, NGLs are included as oil production. 112

World Energy Outlook 2013 | Global Energy Trends

The Middle East is endowed with more convenƟonal gas resources than any other region except for Eastern EuropeͬEurasia, but geƫng the gas to market is proving extremely diĸcult, in part because low regulated prices oŌen discourage investment. The projected increase in the region’s gas producƟon – 305ථbcm, or just under 2% per year, between 2011 and 2035, while substanƟal, represents growth at a slower rate than in Africa or Asia. Qatar is expected to remain the leading contributor to producƟon growth in the region over the projecƟon period, though the recent breakneck expansion in producƟon capacity – most of which serves LNG plants – slows in the near term as the current wave of development projects comes to an end. Qatari gas producƟon is projected to reach a plateau of about 180ථbcm by 2015, but a further increase later in the Outlook period to around 240 bcm in 2035, is projected on the assumpƟons that the moratorium on development of the North Field – part of the world’s largest single gas Įeld, which straddles the border with Iran (where it is called South Pars) – is liŌed and that the government authorises new LNG and GTL projects. With an increasing number of LNG producers entering the market in the next decade, Qatar’s share of the global LNG market is set to fall in the medium term.

© OECD/IEA, 2013

ProducƟon in Saudi Arabia is set to rise in the next few years, with the commissioning of the 30ථbcm per year
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18

Chapter 3 | Natural gas market outlook

113

In LaƟn America, domesƟc needs will drive most upstream developments in the coming decades. In ArgenƟna, the region’s largest producer, shale resources are expected to underpin a revival in gas producƟon, which has been declining in recent years because of dwindling investment – mainly the result of low regulated prices. The potenƟal for producƟon growth is biggest in Brazil, which has discovered large resources of associated gas oīshore and pockets of gas onshore (see Chapter 11). By contrast, producƟon in Trinidad is set to decline over the longer term unless new discoveries are made or the regulatory framework changes. Box 3.3 ‫ ٲ‬Levant gas on the rise Large oīshore gas discoveries in the eastern Mediterranean Sea can change the energy landscape of the whole region. A 2010 assessment by the US Geological Survey esƟmates that undiscovered gas resources in the Levant Basin could amount to 3.5ථtcm – more than six Ɵmes the region’s current proven reserves. Thus far, the bulk of the discoveries made have been oīshore Israel, including the two largest Įelds – Leviathan and Tamar. For Israel, heavily dependent on imported energy, the prospect of developing a large indigenous resource represents a major turn-around. ProducƟon from the Tamar Įeld began in spring 2013 (a very quick development given that the Įeld was discovered only in 2009), helping to compensate for declining output from exisƟng Įelds.

© OECD/IEA, 2013

The Ɵming for development of the larger Leviathan Įeld (with esƟmated recoverable resources of 0.5ථtcm) and other areas of oīshore potenƟal depends on the balance that Israel chooses to strike between reserving gas for the domesƟc market (even though long-term domesƟc needs are uncertain) and sancƟoning export projects. In the projecƟons, producƟon rises steadily from the current low base and approaches 20ථbcm by 2035. This is well in excess of the volumes that could be consumed on the domesƟc market, implying a growing contribuƟon from Israel to gas balances further aĮeld. There are several export opƟons: building domesƟc LNG infrastructure, or ŇoaƟng LNG faciliƟes at the Įelds, would give Israel Ňexibility over export desƟnaƟons͖ alternaƟvely, the gas could be piped to Turkey, Jordan or even Egypt. The oīshore potenƟal in the eastern Mediterranean is by no means conĮned to Israel and there are signs of increasing interest from major gas companies in the region’s resources, following the trail set by the medium-size and independent companies in making the early discoveries. There are sƟll numerous obstacles that may hinder further development of the Levant Basin’s potenƟal, including regional conŇicts, territorial disputes as well as regulatory uncertainty. But, if poliƟcs allow, there is also a realisƟc prospect that gas from this area will soon be making its mark in parts of southeast Europe as well as the increasingly gas-thirsty markets of the Middle East.

114

World Energy Outlook 2013 | Global Energy Trends

&ŽĐƵƐŽŶƵŶĐŽŶǀĞŶƟŽŶĂůŐĂƐ

1

Resources UnconvenƟonal gas will surely play an increasingly important role in future gas supply – not just in North America, the site of the overwhelming bulk of producƟon today, but also in several other parts of the world (Tableථ3.3). Resources of unconvenƟonal gas (shale, Ɵght gas and coalbed methane) are globally abundant (Figure 3.6). Indeed, the Įgures for shale gas that underpin the modelling have been revised upwards for many countries, based on the updated assessment of 137 shale formaƟons in 41 countries (US EIA, 2013).6 But there are numerous obstacles to developing these resources at anything like the scale seen in North America, so replicaƟng that region’s success will be neither easy nor quick. As other countries move down the path towards commercial exploitaƟon of unconvenƟonal gas resources, growing awareness of these obstacles is injecƟng a new realism into discussions about the extent and Ɵming of global prospects for unconvenƟonal gas.

2

Box 3.4 ‫ ٲ‬High-Level Unconventional Gas Forum-towards global best practice

7

The IEA High-Level UnconvenƟonal Gas Forum was created in 2013 to enable governments, industry and other key stakeholders to share insights into best pracƟces in operaƟons, regulaƟons and methods to sƟmulate the widespread and sustainable development of unconvenƟonal gas. The groundwork for the forum was laid by the release of several WEO reports on the subject of unconvenƟonal gas. WEO-2009 highlighted the potenƟally major role that could be played by unconvenƟonal gas worldwide. In 2011, a WEO special report, Are We Entering a Golden Age of Gas?, analysed the global prospects for gas markets, and a subsequent one in 2012, Golden Rules for a Golden Age of Gas, looked speciĮcally at unconvenƟonal gas, oīering guidance for industry and policymakers on how to ensure that its producƟon is conducted in a socially and environmentally acceptable way (IEA, 2011 and 2012).

© OECD/IEA, 2013

The Įrst meeƟng of the Forum was held in March 2013. It was aƩended by 130ථrepresentaƟves from governments, internaƟonal organisaƟons, industry, nongovernmental organisaƟons and investors from all corners of the globe, who shared their experiences regarding unconvenƟonal gas development, including how to deal with social, environmental and economic challenges. It was agreed that best pracƟce regulaƟon should be a parƟcular focus of the Forum’s work. The next meeƟng is planned to be held in the Įrst half of 2014. Diīerences in geological, regulatory and market condiƟons will dictate the nature and pace of development in each region. ProducƟon may be held back by several factors, including: unfavourable geology͖ concerns about the environmental impact of hydraulic fracturing, parƟcularly water management and the risk of methane leaks to the atmosphere (the laƩer reducing the net climate beneĮts of using lower-carbon natural gas as a subsƟtute for coal and oil)͖ local opposiƟon to the disrupƟon caused by drilling (the likelihood of which is 6.ഩ The previous report, released in 2011, covered 69 shale formations in 32 countries.

Chapter 3 | Natural gas market outlook

115

3 4 5 6

8 9 10 11 12 13 14 15 16 17 18

© OECD/IEA, 2013

116

Figure 3.5ථ‫ٲ‬

Remaining unconventional gas resources in selected regions, end-2012 (tcm)

25 20 15 10 5 0

20 15 10 5 0 European Union

Canada

World Energy Outlook 2013 | Global Energy Trends

30 25 20 15 10 5 0 United States

25 20 15 10 5 0

20 15 10 5 0

35 30 25 20 15 10 5 0

45 40 35 30 25 20 15 10 5 0

Russia

China

10 5 0

Algeria

India

Mexico 10 5 0

10 5 0

Indonesia Brazil

Tight gas CBM

25 20 15 10 5 0

15 10 5 0 South Africa

Argentina Shale gas This map is without prejudice to the status of or sovereignty over any territory, to the delimitation of international frontiers and boundaries, and to the name of any territory, city or area.

30 25 20 15 10 5 0

Australia

heightened by systems of mineral rights ownership that, in contrast to much of the United States, do not give surface landowners a stake in the development)͖ lack of access to resources and transportaƟon infrastructure͖ absence of a local oil services industry͖ and unaƩracƟve investment opƟons and regulatory framework. Many of these factors can be inŇuenced directly or indirectly by policy. On the other hand, further advances in producƟon technology could accelerate developments. Global producƟon of unconvenƟonal gas in 2011 is esƟmated to have been around 560ථbcm. In the New Policies Scenario, it rises to 1ථ330 bcm in 2035, its share of total world gas producƟon climbing from 17% to 27%.7 The United States remains the leading unconvenƟonal gas producer, but several countries – notably China – emerge as important producers. Shale gas output rises most in absolute terms, but the output of coalbed methane quadruples (Table 3.5). Inevitably, the uncertainty surrounding these projecƟons is especially acute, given the immaturity of the unconvenƟonal gas sector outside the United States. Since it is WEO pracƟce not to assume radical technological breakthroughs (though technology learning over Ɵme is allowed for), the projecƟons for unconvenƟonal gas output do not include producƟon from methane hydrates (Boxථ3.5).

Scenario 2025

2030

2035

232

402

513

627

745

513

5.0%

78

148

202

261

315

237

6.0%

Tight gas

250

281

285

276

269

18

0.3%

Total

560

832

999

1ථ165

1ථ328

769

3.7%

Shale gas Coalbed methane

Delta

CAAGR*

5 6 7

10

12

WƌŽĚƵĐƟŽŶƚƌĞŶĚƐ More than half of the growth in unconvenƟonal gas producƟon over the period to 2020 is projected to come from the two main established producers, the United States and Canada, which accounted for 90% of total producƟon in 2011. By 2020, their share in global unconvenƟonal producƟon drops to 80%, as producƟon in China and Australia starts to grow. AŌer 2020, the picture becomes much more diverse (Figureථ3.6). The surge in the producƟon of unconvenƟonal gas in the United States, especially shale gas, slowed somewhat in 2012, as very low gas prices led to fewer rigs drilling gas wells. Nonetheless, producƟon of unconvenƟonal dry gas has remained high. There are a number

© OECD/IEA, 2013

4

11

*ථCompound average annual growth rate.

7.ഩ The demarcation between conventional and unconventional gas is not always clear cut, especially with respect to tight gas and conventional gas with reservoir stimulation. We classify tight gas as unconventional according to whether we consider that special production techniques, such as hydraulic fracturing, will be needed for its production. Coalbed methane and shale gas are categorised as unconventional gas in our definition.

Chapter 3 | Natural gas market outlook

3

9

2011-2035

2020

2

8

Table 3.5 ‫ ٲ‬Global production of unconventional gas in the New Policies

2011

1

117

13 14 15 16 17 18

of reasons for this, including the working oī of a backlog of drilled but not completed wells, drilling for dry gas in low cost areas, such as the Marcellus shale, and the producƟon of large volumes of dry gas a by-product of wells targeted at areas of liquids-rich gas, such as the Eagle Ford shale. We nonetheless assume that, over Ɵme, gas prices will move towards a range where producƟon of drier gas is more proĮtable, i.e. within the range of $4.50-6ͬMBtu. At these levels, we judge that large volumes of shale gas can be produced, which is why North American gas prices plateau for a period around these levels before increasing again towards the end of the Outlook period. Total unconvenƟonal gas producƟon in the United States conƟnues to increase in the projecƟons, reaching nearly 600ථbcm in 2035. There is liƩle sign that resource limitaƟons kick in before the end of the projecƟon period, in contrast to our current expectaƟons for US producƟon of light Ɵght oil (see Chapterථ14). Figure 3.6 ‫ ٲ‬Growth in unconventional gas production by type in selected

Shale gas

United States China Canada Argenna India European Union Algeria Mexico Indonesia

CBM

regions in the New Policies Scenario

Australia China India Canada United States

2011-2020 2020-2035

© OECD/IEA, 2013

0

20

40

60

80

100

120

140

160 bcm

Any adverse change in the generally favourable regulatory and operaƟng environment in the United States could have a material impact on the outlook for unconvenƟonal gas producƟon. The likelihood of this is linked, in turn, to the way that the industry meets public concerns about the environmental impact of hydraulic fracturing and other contenƟous aspects of unconvenƟonal gas development. These are, in principle, manageable – as demonstrated by the IEA Golden Rules (IEA, 2012a) – but the industry will need to be vigilant. There are iniƟaƟves underway at the federal level that could inŇuence this picture. The US Environmental ProtecƟon Agency, for example, is preparing an analysis of the impact of hydraulic fracturing on drinking water and ground water, which is due to be released for public comment in 2014. The US Department of Interior has proposed a strengthening of federal regulaƟon of hydraulic fracturing on publicly owned land, in order to establish baseline environmental safeguards for these operaƟons across all public and Indian lands. At the state level, the regulatory picture varies signiĮcantly, so it is diĸcult to draw naƟonwide conclusions. States have the authority to set regulaƟons that apply

118

World Energy Outlook 2013 | Global Energy Trends

to unconvenƟonal gas producƟon as well as their degree of stringency, and have chosen diīerent regulatory approaches, some opƟng for detailed regulaƟon, others for more Ňexible performance standards and case-by-case permiƫng (Richardson, et al., 2013). Where shale gas acƟvity is underway, regulatory changes have tended to Ɵghten the requirements related to well construcƟon and protecƟon of groundwater.8 However, some states maintain or are considering outright prohibiƟons of certain acƟviƟes: New zork and New Jersey have temporarily banned hydraulic fracturing, pending addiƟonal research and data on environmental impacts. Box 3.5 ‫ ٲ‬Are methane hydrates the next revolution-in-waiting? Methane hydrates are deposits of natural gas trapped together with water in a crystalline structure that forms at low temperatures and moderate pressures. They can be found either on the sea Ňoor, in shallow sediments beneath the sea Ňoor or underneath ArcƟc permafrost. Methane hydrates may oīer a future means to further increase the supply of natural gas. Though quanƟtaƟve esƟmates vary by several orders of magnitude, all agree that the resources in place are extremely large, with even the lower esƟmates giving resources larger than all other natural gas resources combined. Many esƟmates fall between 1ථ000 and 5ථ000 tcm, or between 300 and 1ථ500 years of producƟon at current rates. The US Geological Survey esƟmates that gas hydrates worldwide are between 10 to 100 Ɵmes as plenƟful as US shale gas reserves. Producing gas from methane hydrates poses huge technological challenges and the relevant extracƟon technology is in its infancy. So far there have been only smallscale experimental producƟon projects: the Japanese Nankai Trough project has just achieved small-scale producƟon and the Malik project in Canada produced for about three months from one well. The longer-term role of methane hydrates will depend on climate change policies as well as technological advances, as meeƟng ambiƟous goals to reduce emissions would require a reducƟon in demand from all fossil fuels, certainly in the longer term. In addiƟon, methane released to the atmosphere from any source is a potent greenhouse gas and great care has to be taken to minimise such releases – a point highlighted in Redrawing the Energy Climate Map: World Energy Outlook Special Report (IEA, 2013b). One aim of the Japanese research programme is to develop producƟon technology that achieves controlled release of the methane from the ice into the producƟon well, minimising the risk of the methane escaping into the atmosphere. For countries like Japan that conƟnue to rely on expensive imported energy, methane hydrates may be an aƩracƟve energy supply opƟon. The Japanese government aims to achieve commercial producƟon in ten to ĮŌeen years, i.e. by the mid- to late-2020s.

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

© OECD/IEA, 2013

17 8.ഩ An example is the updated requirements related to well casing, cementing, drilling, control and completions that were passed by the Texas Railroad Commission and apply to all wells drilled in Texas from January 2014.

Chapter 3 | Natural gas market outlook

119

18

Figure 3.7 ‫ ٲ‬Unconventional gas production by selected country in the New Policies Scenario Rest of world

1 200

Argenna

1 000

Australia

bcm

1 400

China

800

Canada

600

United States

400 200 2000

2011

2020

2025

2030

2035

© OECD/IEA, 2013

Canada’s unconventional gas production is currently around 70ථbcm, consisting of tight gas along with smaller volumes of shale gas and coalbed methane. Shale gas provides most of the increase that takes unconventional output to 140ථbcm in 2035. Somewhat paradoxically, the boom in shale gas production in the United States has hindered the development of the unconventional gas sector in Canada, as the need for imports from Canada into the United States has slumped, driving down prices across the North American market and depressing incentives to invest in Canada. Nonetheless, there is potential for Canadian LNG exports and companies are now concentrating on exploration in oil and liquids-rich areas of British Columbia (including the Montney play) and Alberta (including the Duvernay and Cardium plays), and the northern part of the Bakken shale, which is predominately an oil play, which extends across the US border into Canada. There is a moratorium in place on hydraulic fracturing in Quebec (where companies had been hoping to develop the Utica shale), which could be extended to allow for further studies on the environmental impact. Mexico is also set to become a signiĮcant producer of unconvenƟonal gas in the longer term, with output reaching almost 30ථbcm by 2035 in the New Policies Scenario. Pemex, the naƟonal oil and gas company with a monopoly over all upstream developments, has launched a $200 million three-year programme to explore for shale gas in Mexico, starƟng with the extension into the north of the country of the Eagle Ford play, which is thought to hold close to half of the country’s shale resources. But commercial producƟon may be constrained by: water scarcity in some of the resource-rich areas͖ the absence of rights for companies other than Pemex to work in the upstream sector͖ the priority given by Pemex’s investment in exportථrevenue-generaƟng oil projects͖ and the diĸculty in keeping development costs at a level that allows compeƟƟon with imported gas from the United States. Prospects though could brighten if proposals for reform of the country’s oil and gas sectors were to bring new investment capital. Europe is well-endowed with all three types of unconvenƟonal gas, but large-scale development must deal with geological condiƟons that are considered to be more diĸcult 120

World Energy Outlook 2013 | Global Energy Trends

than those in North America, as well as public and poliƟcal opposiƟon in many countries to unconvenƟonal gas producƟon, especially in Western Europe. It is for the moment unclear to what extent these social and environmental concerns will Įnd reŇecƟon in a Ɵghtening of the regulatory framework at European level. In the New Policies Scenario, we take a generally cauƟous view of the prospects for producƟon, which reaches 20ථbcm by 2035 in the European Union. The largest share of this (8ථbcm) comes from Poland, which has been seen as the most promising country for unconvenƟonal producƟon. As of September 2013, more than 50 wells have been completed (and another 200 or so are planned by 2016). Thus far the results have not met the industry’s iniƟal expectaƟons, although it is sƟll early to make judgement on the scale or quality of exploitable resources (only seven horizontal wells have undergone mulƟ-stage fracturing). The United
© OECD/IEA, 2013

In Australia, coalbed methane has been the main focus of unconvenƟonal gas development and producƟon is set to climb steeply with the compleƟon of three LNG plants that are being built in Gladstone in Queensland, to be fed by natural gas from coal seams in the Surat Basin. In the projecƟons, coalbed methane producƟon in Australia rises from just 6ථbcm in 2011 to almost 100ථbcm by 2035. For this to be realised, operators will need to pay parƟcular aƩenƟon to water management: water is a parƟcularly sensiƟve maƩer in Australia, given general water scarcity and the high reliance in some regions on ground and artesian water sources for agricultural and grazing acƟvity. As in most federal systems, environmental regulaƟon is mostly a state responsibility, with regulaƟon focused on guaranteeing well integrity and management of the large quanƟƟes of formaƟon water that must be removed prior to gas producƟon. A warning shot for the industry was the decision in New South Wales in early 2013 to ban coalbed methane developments within two kilometres of residenƟal areas or certain rural acƟviƟes, causing at least one major development to be suspended in that state and exploraƟon acƟviƟes to be curtailed. Subsequently, in March 2013, the federal government indicated that federal approvals for new coalbed methane projects (and large coal mining developments) will be required where they signiĮcantly impact water resources. In China, coalbed methane is already in commercial producƟon, with 10ථbcm marketed in 2011. But producƟon is rising less rapidly than planned and the target of 30 bcm by 2015 that was set in the 12th Five-zear Plan is unlikely to be met͖ in our projecƟons, coalbed methane producƟon reaches 30ථbcm only closer to 2020. The potenƟal for shale gas producƟon in China is much larger, but projects are sƟll largely at the exploraƟon stage. Two licensing rounds have been completed, mainly in the Sichuan region, with foreign companies allowed to parƟcipate as minority partners of a Chinese company. In at least two instances the foreign company is operator, with important implicaƟons for technology

Chapter 3 | Natural gas market outlook

121

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18

transfer. Shell and Hess have already signed producƟon-sharing agreements. However, the anƟcipated increase in commercial producƟon is unlikely to get China close to the oĸcial target of 6.5ථbcm of shale gas producƟon by 2015. In the projecƟons, China’s shale gas producƟon builds up relaƟvely slowly unƟl the laƩer part of this decade, but then accelerates as the industry starts to scale up its eīorts, to reach almost 120ථbcm per year by 2035. The increase is facilitated by gradual changes in gas pricing policy, which help to improve incenƟves for exploraƟon and development of unconvenƟonal gas. The resource base could probably support a considerably higher level of producƟon than we project, but producƟon is expected to be held back by several factors. There are sƟll signiĮcant quesƟons over how favourable the geology is (for example, the gas-bearing formaƟons are in most cases much deeper than in North America, making them potenƟally more costly to develop). Some of the most promising resources are also located in mountainous areas where access is diĸcult. Limited availability of water, parƟcularly in the Tarim and Ordos Basins, and of pipeline and processing capacity could also hinder development.

© OECD/IEA, 2013

In India, there is currently no unconvenƟonal producƟon, but its value as a means to meet growing demand and reduce import dependence is gaining recogniƟon. Shale gas appears to be the most promising prospect and the government is developing speciĮc rules for the exploraƟon and development of shale gas ahead of a planned aucƟon of licenses for at least 100 blocks. The aucƟon has been delayed while the environment ministry completes studies on the implicaƟons of hydraulic fracturing for both the availability and quality of water supplies. Land acquisiƟon rights have also complicated maƩers: local protests have occurred over the use of farmland in West Bengal for shale gas drilling by the state-owned Oil and Natural Gas CorporaƟon. For commercial producƟon to proceed in the event of successful drilling, transport infrastructure will need to be built and the Įscal regime adapted. Prices have been too low to make shale gas producƟon proĮtable, but a government decision in June 2013 to raise gas prices is expected to boost interest in drilling. The government also plans to introduce a scheme whereby the states would receive a 10% royalty on producƟon, similar to the one already being used for country’s coalbed methane. In the New Policies Scenario, commercial producƟon of shale gas and coalbed methane starts towards the end of the current decade, with shale gas output rising to almost 35ථbcm by 2035 and coalbed methane to 25ථbcm. Indonesia has also been pushing ahead with plans to develop its unconvenƟonal gas resources and, like India, is projected to produce both shale gas and coalbed methane from the 2020s, with combined output of around 20ථbcm by 2035. Five companies have Įnished a joint study regarding shale gas potenƟal in North Sumatra and around 70 proposals to drill exploraƟon wells have been submiƩed for approval, following a Įrst licensing round in the area. Licensing rounds for other prospecƟve areas are planned in the coming months. The government expects commercial shale gas producƟon to begin in 2018. ExploraƟon acƟvity is also underway for coalbed methane and dozens of producƟon-sharing agreements have been signed. The regulatory regime for unconvenƟonal gas, including the sharing of competences between local and central government, is under development, with tax incenƟves planned to bring forward investment. 122

World Energy Outlook 2013 | Global Energy Trends

Algeria’s interest in unconvenƟonal gas is noteworthy as it comes from a major convenƟonal producer with signiĮcant remaining convenƟonal resources looking to address the problem of declining output. Amendments to hydrocarbon legislaƟon in 2013 introduced tax incenƟves for shale and Ɵght gas development, but it remains to be seen whether these will be suĸcient to bring projects to fruiƟon, parƟcularly in the light of concerns over security and the scarcity of water resources. In the New Policies Scenario, Algerian unconvenƟonal output (almost all of which is shale gas) gains momentum only towards the end of the projecƟon period, reaching 15ථbcm in 2035. In the recent US EIA assessment, ArgenƟna is ranked second in the world for shale gas resources (US EIA, 2013). The most promising play is Vaca Muerta in northern Patagonia. The geological prospects for producƟon appear posiƟve, but Įscal, contractual and poliƟcal obstacles are expected to slow development in pracƟce. Companies are expected to focus on oil and liquids-rich areas in preference to drier gas resources. One of the factors holding back investment in shale plays has been the low price oīered for producƟon. This was addressed with a government decision in February 2013 to triple the wellhead price for all types of gas to $7.50ͬMBtu. zPF – the newly naƟonalised leading producer in ArgenƟna – has a $6.5ථbillion capital expenditure programme for gas that aims to boost overall producƟon by 8% per year over 2013-2017, with about 60% of the incremental producƟon coming from Ɵght and shale gas. It has also announced partnerships with local and internaƟonal companies to develop ArgenƟna’s resources, including a joint-venture agreement with Chevron to develop the Vaca Muerta Įeld (though the deal is being contested by Repsol, the former owner of zPF). On the assumpƟon that these deals bear fruit, shale and Ɵght gas producƟon reaches more than 50ථbcm per year by 2035 in the New Policies Scenario in addiƟon to almost 40ථbcm from convenƟonal resources.

Trade, pricing and investment Inter-regional gas trade has risen by 80% over the last two decades and we project that it conƟnues to follow an upward path in the New Policies Scenario, expanding by some 400ථbcm to reach 1ථ090ථbcm in 2035 (Table 3.6).9 This promises to be a very dynamic period for internaƟonal trade in gas, with the rising importance or emergence of strong new market players, notably Australia, the United States, Canada and countries in East Africa, who provide a compeƟƟve challenge to established exporters such as Russia and Qatar. The period also sees a conƟnued shiŌ in the direcƟon of internaƟonal gas trade, away from the AtlanƟc basin (although Europe conƟnues to be the largest single imporƟng region) and towards the Asia-PaciĮc region, a shiŌ that poses new dilemmas for Eurasian producers reliant on Įxed pipeline infrastructure for access to market. And there are signs that the terms of internaƟonal trade – parƟcularly in the form of LNG – will become more © OECD/IEA, 2013

2 3 4 5 6 7 8 9 10 11 12

Inter-regional trade

9.ഩ Inter-regional trade is trade between the major WEO regions͖ its rise has been interrupted twice in the last four years – the first time by the recession-induced fall in demand in 2009 and the second in 2012 because of continuing weak import demand in Europe and a decline in LNG supply.

Chapter 3 | Natural gas market outlook

1

123

13 14 15 16 17 18

sensitive to short-term market conditions, with innovative pricing and fewer destination clauses, bringing new connections between the different regional markets and changes in the way gas is priced around the world. Table 3.6 ⊳ Net natural gas trade by region in the New Policies Scenario (pipeline and LNG) 2011

OECD Americas United States

2020

Trade (bcm)

Share of demand or output (%)*

-402 -11 -47

2035

Trade (bcm)

Share of demand or output (%)*

Trade (bcm)

Share of demand or output (%)*

25%

-349

20%

-402

21%

1%

43

4%

69

6%

7%

15

2%

48

6%

Europe

-248

47%

-288

54%

-390

65%

Asia Oceania

-143

71%

-105

49%

-81

34%

-117

97%

-117

99%

-123

100%

Non-OECD

415

19%

351

13%

407

12%

E. Europe/Eurasia

179

20%

179

20%

347

30%

Caspian

58

33%

76

37%

143

50%

Russia

197

29%

174

26%

263

33%

9

2%

-103

15%

-319

29% 40%

Japan

Asia China

-29

22%

-130

42%

-212

India

-14

24%

-25

29%

-74

43%

120

23%

119

19%

123

15%

Middle East Africa

89

44%

127

45%

224

52%

Latin America

19

11%

29

13%

32

10%

Brazil

-10

38%

-7

16%

2

2%

World**

685

20%

804

20%

1 092

22%

-308

63%

-360

73%

-450

81%

European Union

* Imports as a share of primary demand for importing countries; exports as a share of production (output) for exporting regions/countries. ** Total net exports for all WEO regions, not including trade within WEO regions. Notes: Positive numbers denote exports; negative numbers imports. The difference between OECD and non-OECD totals in 2011 is due to stock change and statistical discrepancies.

Despite a relatively modest increase in demand, Europe’s need for imported gas grows more strongly over the projection period, as production falls back across the continent (Norway is the important exception). In the case of the European Union, the gas import requirement rises by some 140 bcm to reach 450 bcm by 2035 (Figure 3.8). Europe is wellplaced to secure this supply from a variety of sources, including traditional suppliers such as Norway (which became the European Union’s largest single supplier of gas in 2012), Russia and Algeria, as well as from the international LNG market. There are also newcomers looking to supply Europe by pipeline, notably Azerbaijan and, potentially, also Iraq, along the so-called “southern corridor” through Turkey and the rest of southeast Europe. The Asia-Pacific region is the arena in which the most profound changes in global gas markets are set to play out over the coming decades, though the speed and extent of 124

World Energy Outlook 2013 | Global Energy Trends

those changes is subject to a high degree of uncertainty. Outside Japan and
2

Figure 3.8 ‫ ٲ‬European Union natural gas supply and demand balance in the

4

bcm

New Policies Scenario 600

3

5

Net imports Producon

400

1

Consumpon

6

200

7

0

8

-200 -400 -600 1990

9 2000

2010

2020

2030

2035

© OECD/IEA, 2013

Japan,
125

10 11 12 13 14 15 16 17 18

to 47ථbcm in 2035 from 17ථbcm today, and also, potenƟally, from other countries in the region, notably from Iraq. The volumes delivered along this corridor remain relaƟvely small compared with European gas demand, but nonetheless promise to make a contribuƟon to diversity and security of supply. Figure 3.9 ‫ ٲ‬China natural gas supply and demand balance in the New

bcm

Policies Scenario 600

Net imports

500

Consumpon

400

Producon

300 200 100 0 -100 -200 -300 1990

2000

2010

2020

2030

2035

The volume of Russian pipeline exports rises only modestly over the period to 2020, despite the possible addiƟon of new export capacity in the form of the South Stream pipeline or addiƟonal North Stream pipelines. More rapid growth is assumed to be constrained by Russia’s stance on pricing gas in Europe, where the defence of oil-indexed pricing clauses could come at the expense of market share. AŌer 2020, however, Russian pipeline exports are expected to expand once again as its focus shiŌs to the east, with most of the growth coming from an assumed new link to China from Russia’s east Siberian Įelds. China is also expected to draw increasing volumes of gas from Central Asia, where the exisƟng pipeline link from Turkmenistan is assumed to expand to 60ථbcm of capacity, and from Myanmar, with which a 12ථbcm per year link was completed in 2013.

© OECD/IEA, 2013

Of the overall increase in inter-regional gas trade of more than 400ථbcm, pipelines carry just under half. The slightly larger share (210ථbcm) is expected to come in the form of LNG. Whereas pipeline trade remains dominated by a few producers, primarily in Eurasia, the ranks of LNG exporters are set for a major re-shuŋe. Among some of the exisƟng exporters, there are already signs of rapidly increasing domesƟc demand limiƟng supply for export. This trend is most notable in the Middle East, where Qatar and zemen may become the sole remaining LNG exporters by the early 2020s (although they may be joined later by Iraq andͬor Iran).10 Egypt and Trinidad and Tobago are among other current exporters which may have to cut back external commitments.

10.ഩ Oman’s sales and purchase agreement for LNG exports expires by 2024 and may not be renewed because of booming local demand and insufficient gas supplies. The United Arab Emirates will need to decide whether to do likewise when the current contract to supply LNG to Japan from a plant in Abu Dhabi terminates in 2019͖ Abu Dhabi is already a net gas importer. 126

World Energy Outlook 2013 | Global Energy Trends

Box 3.6 ‫ ٲ‬How great should expectations be for North American LNG?

1

The emergence of North America as an LNG exporter is based in part on proposed greenĮeld projects in the United States and western Canada, but the Įrst new projects in the United States are based rather on conversion of terminals that were originally intended to handle LNG imports. The construcƟon of liquefacƟon faciliƟes is expensive, but exisƟng terminals have shipping and storage infrastructure already in place, reducing the overall cost and the Ɵme required to get projects operaƟonal. As of October 2013, four such projects have thus far received condiƟonal approvals from the US Department of Energy allowing them to move ahead.11 These include: Sabine Pass and Lake Charles projects in Louisiana͖ Freeport project in Texas͖ and the Cove Point project in Maryland, which will provide annual export capacity totalling some 65ථbcm. Sabine Pass has also received the necessary supplementary approval from the USථFederal Energy Regulatory Commission to construct the liquefacƟon faciliƟes and is expected to begin operaƟon in 2016. As of October 2013, there are an addiƟonal 28 applicaƟons to export LNG from the United States at various stages of the approvals process, together providing for a theoreƟcal addiƟon of more than 250ථbcm per year in export capacity. Only a fracƟon of these are expected to see the light of day, but a few are nonetheless advancing, with the Cameron project in Texas already having long-term export commitments in place with prospecƟve buyers. The business model for US LNG export projects, at least for the iniƟal projects, is disƟncƟve by internaƟonal standards. Instead of being supported by long-term contracts, with pricing linked to the oil price and exports dedicated to a single desƟnaƟon, they are based on the Henry Hub price, plus a liquefacƟon fee, and there are no desƟnaƟon restricƟons. The tolling (liquefacƟon) fee is set by long-term contract. The net result is that this LNG is eīecƟvely free to seek the most advantageous internaƟonal market. In most cases, this is expected to be in Asia.

© OECD/IEA, 2013

In Canada, the focus for LNG export projects is largely on the west coast. Of the seven proposed projects in BriƟsh Columbia, three have received export licenses and are at the stage of seeking environmental approvals: Shell’s LNG Canada (with ulƟmate capacity of 32ථbcm per year), Chevron’s
11.ഩIn the case of the United States, export approvals are required from the Department of Energy (DoE), a routine matter for applications to export to countries with which the United States has a free-trade agreement, but a more lengthy process – involving an assessment of the public interest – for export to other countries. Facility approvals are also required from the Federal Energy Regulatory Commission. Overall, the approval process can take at least two years and cost $100 million or more. At a federal level in Canada, all projects must be approved by the National Energy Board.

Chapter 3 | Natural gas market outlook

127

2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18

On the other hand, new LNG exporters are also set to emerge, while some exisƟng ones expect to see their market shares rise. Worldwide, there are twelve LNG export plants under construcƟon today with a combined capacity of around 130ථbcm per year.12 This new capacity is set to come into operaƟon between 2015 and 2018, although the Ɵmetable is heavily conƟngent on what happens in Australia, where seven of the twelve terminals are located and where projects have seen cost escalaƟons and delays. AŌer Australia, the next new tranche of LNG supply is set to come from North America (Boxථ3.6). With producƟon outstripping domesƟc demand, by 2035 net exports from the United States reach almost 50ථbcm and 45ථbcm from Canada (pipeline and LNG), with net North American LNG exports as a whole reaching around 50ථbcm by 2020 and 75ථbcm by the end of the projecƟon period.13 These projecƟons are highly sensiƟve to changes in the outlook for demand and producƟon – relaƟvely small shiŌs in either could have a large impact on the overall trade balance. There is potenƟal upside to these Įgures (parƟcularly those from the United States) that would have implicaƟons for other exporters around the world – a possibility that we examine in more detail below in a Gas Price Convergence Case. The rise of LNG export from Australia and from North America is accompanied, in the projecƟons, by new projects, based on oīshore developments in East Africa, as well as by expansion of capacity among some exisƟng LNG exporters, including Russia. The Russian expansion may take on added signiĮcance if, as seems possible, RosneŌ and Novatek secure rights to export LNG directly to Asian markets, marking the Įrst major breach in Gazprom’s export monopoly. Over the projecƟon period, higher assumed import prices into the Asia-PaciĮc region make this the desƟnaƟon of choice for most LNG exporters, with Europe assuming the role of balancing the market.

WƌŝĐŝŶŐŽĨŝŶƚĞƌŶĂƟŽŶĂůůLJƚƌĂĚĞĚŐĂƐ

© OECD/IEA, 2013

The way that gas is priced in internaƟonal trade has undergone substanƟal changes in recent years. An iniƟal switch, noƟceable in the period from 2005 to 2008, reduced the share of gas sold under prices established by bilateral negoƟaƟon between large buyers and sellers (Figure 3.10)͖ this was largely due to a switch in pricing policy for Russian gas exports to neighbouring countries (as in the disputed case of Russian exports to Ukraine, this was eventually replaced by pricing based on oil-price indexaƟon).14 But a second change has been the rising share of gas traded under prices set by gas-to-gas compeƟƟon, i.e. prices determined by the interplay of gas supply and demand. In 2005, this share was around one-ĮŌh͖ by 2012, it is esƟmated to have risen to 37%.

12.ഩ This is offset in part by the anticipated decommissioning of some 13 bcm per year of existing LNG capacity in Indonesia, Algeria and Alaska. 13.ഩ These net figures take into account Canadian exports by pipeline to the United States, and pipeline exports from the United States to Mexico. 14.ഩ Trade with prices set by bilateral negotiation is now largely confined to two routes͖ Russia to Belarus and Qatar to the United Arab Emirates. 128

World Energy Outlook 2013 | Global Energy Trends

Figure 3.10 ‫ ٲ‬Estimated shares of internationally traded gas by type of

1

pricing mechanism 100%

2

Bilateral negoaon Gas-on-gas compeon

80%

Oil-price indexaon

3

60%

4

40%

5

20%

2005

2009

6

2012

Source: InternaƟonal Gas Union (2013).

7

The shift towards prices set by gas-to-gas competition has been concentrated in continental Europe. A combination of weak demand resulting from economic contraction, surplus take-or-pay obligations and the short-term availability, at competitive prices, of LNG no longer required in North America put the system of oil indexation in long-term supply contracts under substantial pressure. The result of negotiation and, occasionally, arbitration between European importers and their external suppliers has been a higher share of gas sold with reference to the prices at European gas trading hubs, lower base prices, as well as revisions to take-or-pay provisions. By most estimates, the share of gas sold under oil indexation has already fallen towards half of the gas sold in Europe, although the boundary between prices set by oil indexation and those established by gas-to-gas competition is not a precise one.15 There are also marked regional variations, with oil-indexed gas dominant in the south of the continent, but – with a share of less than 30% – increasingly rare in the northwest.

© OECD/IEA, 2013

In the Asia-PaciĮc region, there have been fewer signs of change to the pricing terms for the bulk of regional trade, which conƟnues to be based overwhelmingly on oil-indexed long-term contracts (many with clauses restricƟng the potenƟal for re-sale of the gas). The enduring role of this type of contract reŇects a premium that buyers have been obliged to pay for security of supply, the ability of regulated uƟlity buyers to pass the addiƟonal costs on to their customers and – by way of contrast to the condiƟons prevailing in Europe – the region’s relaƟvely Ɵght gas markets, which meant that sellers could generally set the terms of sale.

15.ഩ In practice, oil, gas (and power) price indices can combine in a variety of ways to determine the level and movement of gas prices in long-term contracts. The International Gas Union (2013) puts the share of European gas imported under oil indexation at 60% in 2012, with the remainder set by gas-to-gas competition.

Chapter 3 | Natural gas market outlook

129

8 9 10 11 12 13 14 15 16 17 18

With the projecƟons of robust gas demand and relaƟvely high costs of supply, there would be strong underpinning to prices for gas imports to the Asia-PaciĮc region even under an alternaƟve gas-to-gas system of price formaƟon, but buyers in this region are nonetheless showing increasing interest in diversifying pricing away from oil indexaƟon. The current situaƟon has resulted in import prices for the region that are the highest in the world and many Ɵmes higher than wholesale prices in the United States, undermining the ability of gas to compete with other fuels, burdening the region’s economies with high import bills and raising concerns about industrial compeƟƟveness (see Chapter 8). Over the projecƟon period, we anƟcipate conƟnued momentum behind changes to the pricing of internaƟonally traded gas, with greater reliance over Ɵme on mechanisms that reŇect the supply-demand balance for gas itself, rather than the price of oil. In Europe, at least, the clear trend is towards more widespread adopƟon of hub-based pricing, more use of spot trading and shorter duraƟon of long-term contracts. In the Asia-PaciĮc region, too, we expect alternaƟves to oil indexaƟon to gain ground and the contracƟng structure to become more Ňexible, albeit at a slower pace. Alongside the desire of gas buyers to seek more advantageous terms for their purchases, a catalyst for these changes is the increasing quanƟty of LNG that is set to be available without commitment to a speciĮc desƟnaƟon, loosening the current rigidity of LNG contracƟng structures. As this LNG is free to seek the most advantageous sales price at any given moment, it has the potenƟal to arbitrage price diīerences between markets, increasing the depth and liquidity of short-term LNG trade.16 Such volumes are set to grow, not only from the United States, but also from other projects, notably in East Africa, where a part of the gas is set to be absorbed into the porƞolios of major LNG marketers.17 The desƟnaƟon of other volumes, from North America and East Africa, appear to be preordained, in that they belong to large Asian buyers (
© OECD/IEA, 2013

The result is to create new linkages between regional markets and new interacƟons between their pricing mechanisms. The speed at which change takes place is highly uncertain, depending on market circumstances (that may align in some periods to favour a faster shiŌ, as has been the case in Europe since 2009) and the implementaƟon of policies allowing for regional pricing signals to emerge. The laƩer include third-party access to infrastructure and re-gasiĮcaƟon terminals, the development of compeƟƟve wholesale gas markets and independent regulaƟon (IEA,ථ2013b).

16.ഩ Short-term LNG trade typically includes contracts with duration of up to four years. 17.ഩ Major LNG marketers such as ENI and BG have stakes in the East African gas projects, and would be ideally located to arbitrage between the Atlantic and Pacific basins, providing some competition with Qatar. 130

World Energy Outlook 2013 | Global Energy Trends

With some convergence in pricing mechanisms over the projecƟon period, we expect to see some narrowing of the pricing boundaries within which global trade takes place, i.e. some convergence in gas price levels (see Chapter 1). However, there are limits to the emergence of a fully integrated global gas market in the New Policies Scenario͖ as the regional prices for this scenario show, diīerences in the various gas prices persist – notably for the price of Asia-PaciĮc imports – that are, arguably, above the costs of transporƟng gas (including liquefacƟon and regasiĮcaƟon costs in the case of LNG) between regions (Boxථ3.7). That regional markets remain segmented in the New Policies Scenario reŇects strong forces of inerƟa within the exisƟng system. An underlying reason is the high capital cost of gas infrastructure, which, in many markets, has fostered the use of long-term capacity reservaƟon contracts as a means to reduce risks and lower the cost of capital.18 These contracts have the accompanying eīect of leaving liƩle room for the spot transacƟons that are essenƟal if a global price is to emerge. Seeking to change this situaƟon runs into a chicken-and-egg problem: in regions without an eĸcient trading market, long-term oil-indexed contracts are a logical choice to get projects oī the ground͖ but this has the eīect of hindering the development of markets (perpetuaƟng a lack of conĮdence among producers in the reliability of the price generated through trading markets). Around 80% of the LNG from the twelve projects under construcƟon worldwide has already been contracted on a long-term basis and (with the excepƟon of gas from the single US project among the twelve – Sabine Pass) all of this gas has been sold under contracts with oil indexaƟon. As described, LNG exports from the United States have the potenƟal to change this situaƟon, but, in the New Policies Scenario, the room for their expansion is constrained by prevailing market rigidiƟes and the consideraƟon that a potenƟal second wave of US export projects, aŌer the iniƟal plants based on conversion of LNG import terminals, are greenĮeld developments that would face higher costs. As such, in the New Policies Scenario, these exports from the United States provide a welcome means for Asian buyers to diversify their import pricing structures and sources of supply, but not the basis for a more wide-ranging transformaƟon of the pricing foundaƟons of regional gas trade.

© OECD/IEA, 2013

In considering the possibility of a far-reaching overhaul of pricing structures, it is worth factoring in the staunch resistance from other producers to a change in the way

18.ഩ Capital intensity does not preclude a business model based on sales to a spot market, if market efficiency reaches a critical level. As demonstrated in North America and in projects serving the U< market, deep and liquid gas markets can provide adequate security for large investments in new production or transportation capacity. The development of gas futures markets could also provide a mechanism to lower the risk associated with largescale, long-term gas investments. In the case of LNG, due to their balance sheet and risk-taking ability, the major oil and gas companies can act as anchor consumers for at least a portion of a project and, instead of delivering to captive end-users with a destination clause, they take gas into their global portfolio, with the view of selling it in the most attractive market. zemen LNG, for example, applied this model with Total as the lead shareholder and offtaker.

Chapter 3 | Natural gas market outlook

131

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18

that their exports are priced. Algeria’s Sonatrach and Russia’s Gazprom are at the fore of this opposiƟon (and the summit of the Gas ExporƟng Countries Forum in Russia in June 2013 commiƩed to defending oil indexaƟon). There are also the policy challenges associated with introducing a more compeƟƟve model for gas markets and supply in key Asian countries͖ this was a lengthy and complex process in North America and has been even more so in conƟnental Europe (despite the spur provided by European Union law and compeƟƟon policy). Although individual countries may choose to move quickly, the absence of authoritaƟve supranaƟonal energy policy co-ordinaƟon in Asia gives reason to assume that this process for the region as a whole will be slow. All of these factors serve to put a brake on global gas market integraƟon in the New Policies Scenario.

A Gas Price Convergence Case Nevertheless, given the increasing role of spot trading and rising interconnecƟons between regional markets, it is reasonable to invesƟgate the condiƟons under which convergence between regional pricing mechanisms and prices could be more pronounced than in the New Policies Scenario, and examine the potenƟal implicaƟons for markets, gas demand and trade Ňows of such a development. We discuss these possibiliƟes in an illustraƟve Gas Price Convergence Case, in which the diīerent regional gas markets make a more rapid transiƟon to the point at which they all respond to a single global price signal. At this point, diīerences in regional prices narrow to reŇect only the cost of moving gas between them.

© OECD/IEA, 2013

This case rests on three main conditions that differentiate it from the New Policies Scenario. The first is a larger volume of LNG export from North America (primarily from the United States), which exceeds 100ථbcm by the latter part of the 2020s, more than double the volumes envisaged in the New Policies Scenario. The second condition is that new supply contracts, whether completely new or replacing expiring contracts, are hubpriced in Europe and, even if partly oil-indexed in Asia, not indexed to the traditional JCC mechanism (the average price of crude oil imports to Japan, or Japan Crude Cocktail). This development is accompanied, in the Asia-Pacific region, by an accelerated pace of regulatory change in the gas sector, so as to increase market liquidity and competition among suppliers, including more rapid progress with setting up regional trading hubs that facilitate the exchange of gas.19 A third condition is some easing of costs of constructing liquefaction plants and of LNG shipping, in order to keep down the costs of moving gas between markets (Box 3.7).

19.ഩ Although there are moves to expand gas trading in Singapore and elsewhere in the region, China can play a special role in Asia-Pacific price formation, because it has meaningful potential for both domestic upstream and pipeline imports, as well as LNG imports. With a well functioning gas market based on third-party access to Petrochina’s pipelines, the upstream value which would emerge in Inner Mongolia or Sichuan would be around $8-10ͬMBtu, which, even with all the geological and project management difficulties, would provide a powerful incentive for unconventional gas development. With a competitive domestic upstream, pipeline imports and several LNG terminals, China has the necessary conditions to create a liquid, diversified gas trading hub, possibly developing the current pilot project in Shanghai. 132

World Energy Outlook 2013 | Global Energy Trends

Box 3.7 ‫ ٲ‬Price differentials between regions in a “global” gas market20

1

Movement in the direcƟon of a global gas market does not imply the emergence of a single global gas price, as is largely the case for oil. The key reason for this is the high cost of transporƟng (and storing) gas, related in turn to a much lower energy density than oil. Transforming natural gas into LNG solves the energy density problem, creaƟng a commodity that can be moved more easily and Ňexibly between markets, but comes at the considerable price of construcƟng liquefacƟon plants, specialised LNG carriers and re-gasiĮcaƟon faciliƟes. So, even in a fully ͞converged͟ world of gas prices, there would sƟll be substanƟal diīerences in price between the US Henry Hub price and the respecƟve import prices in Europe and Japan. In the Gas Price Convergence Case, these diīerenƟals narrow to $4.5ͬMBtu between Henry Hub and the European import price, with an extra $1ͬMBtu for imports to Japan, to reŇect the addiƟonal distance. These Įgures are at the low end of our esƟmates for the various components of inter-regional LNG transportaƟon in 2020 (Tableථ3.7). Price diīerenƟals of this type would require that North American projects escape the sort of cost inŇaƟon that has beleaguered LNG developments in some other parts of the world (the large North American market for engineering and contracƟng services gives some cause for opƟmism on this point) and that the United States realises major cost savings by making use of exisƟng infrastructure, including the pipelines, storage and loading faciliƟes at exisƟng regasiĮcaƟon terminals. On the shipping side, costs at the low end of the range shown in Table 3.8 would require a new wave of investment in LNG tankers, bringing charter rates down towards our esƟmate of long-run marginal costs.20 Table 3.7ථ‫ ٲ‬Indicative range of cost estimates for conversion and interUS to Europe

© OECD/IEA, 2013

US to Japan

High

Low

LiquefacƟon

3.0

4.5

3.0

4.5

1.0

2.5

2.0

3.5

RegasiĮcaƟon

0.3

0.5

0.3

0.5

Total

4.3

7.5

5.3

8.5

5 6 7 8 9 10

13 14

These cost esƟmates compare with gas price diīerenƟals between Henry Hub and the European import price in 2035 in the New Policies Scenario of $6ͬMBtu͖ and around $8ͬMBtu between Henry Hub and the Japanese import price. With a pessimisƟc view on the evoluƟon of LNG liquefacƟon and shipping costs (at the high end of this range), it could be argued that the New Policies Scenario already represents a case in which prices have largely converged. This serves to underline that developments in future trade paƩerns and pricing depend not only on developments in producƟon, but also on the way that costs evolve all along the LNG value chain. 20.ഩResponding to today’s high shipping charter rates, almost 120 LNG ships are under order as of September 2013͖ this compares with just over 20 ships on order at the start of 2011.

Chapter 3 | Natural gas market outlook

4

12

High

Shipping

3

11

regional transportation for LNG, 2020 (2012 dollars per MBtu)

Low

2

133

15 16 17 18

A global price signal could emerge based on the spot market for LNG, but in the Gas Price Convergence Case we consider that it comes instead from Henry Hub, which becomes a reference point for global pricing that is transmiƩed to the various regional gas markets via LNG trade. Convergence in prices is assumed to be largely complete by the mid-2020s, i.e. over the next ten years (although it takes longer to complete fully in the case of prices in the Asia-PaciĮc region). Because of higher levels of LNG exports from the United States, the Henry Hub price is higher than it is in the New Policies Scenario (the price impact here is consistent with the Įndings of the study commissioned by the US Department of Energy on the eīect of increased natural gas exports on domesƟc energy markets ΀EIA, 2012΁). The average price of gas imported to Europe falls to $11ͬMBtu in the mid-2020s, before rising in line with changes in the Henry Hub price, while the Japanese import price falls to around $12ͬMBtu over the same Ɵmeframe, remaining around this level before edging slightly higher aŌer 2030 (Figure 3.11). Figure 3.11 ‫ ٲ‬Regional gas prices in the New Policies Scenario and in the 18

NPS

15 12 GPCC

Dollars per MBtu

Gas Price Convergence Case

9 6

New Policies Scenario (NPS): United States Europe Japan Gas Price Convergence Case (GPCC): United States Europe Japan

3

© OECD/IEA, 2013

2000

2005

2010

2015

2020

2025

2030

2035

The prospect of a signiĮcant wave of supplies from new LNG capacity coming onstream in the early to mid-2020s is assumed to strengthen the hand of buyers during contract negoƟaƟons, precipitaƟng a shiŌ in the pricing paradigm in favour of greater Ňexibility and of indices beyond (or in addiƟon to) oil indexaƟon. This provides a boost to inter-regional trade, which, in the Gas Convergence Case, is 30ථbcm higher than in the New Policies Scenario by the mid-2020s and 60ථbcm higher by 2035, all in the form of LNG. A greater share of this LNG arriving on the market (including not only North American, but also East African volumes) is not bound by contract to a speciĮc market. Market and regulatory policies are assumed to be in place, including in the premium Asia-PaciĮc market, to allow for the growth of short-term gas trading and the emergence of transparent regional prices based on gas-to-gas compeƟƟon. This would replicate, to a degree, the gradual transformaƟon of conƟnental European systems of gas price formaƟon which started in the late 2000s͖ but – in this case – it would contribute to a wider process of interconnecƟon between all the major regional gas markets. 134

World Energy Outlook 2013 | Global Energy Trends

What are the implicaƟons of this Gas Price Convergence Case for gas consumpƟon? As one might expect, lower prices sƟmulate extra demand, with global gas use 107ථbcm (2.1%) higher by 2035 than in the New Policies Scenario (Figureථ3.12). Some of this addiƟonal demand is related to the increased compeƟƟveness of gas versus other fuels. Coal consumpƟon grows more slowly in this Case than in the New Policies Scenario (it is 46ථmillion tonnes of coal equivalent lower by 2035), although this eīect is muted because a conƟnued gap persists in most markets between gas and coal prices for power generaƟon. A larger part of the addiƟonal demand for gas stems from the assumpƟon that cheaper gas sƟmulates more aggressive acƟon on issues such as local polluƟon. This facilitates faster growth among emerging Asian economies in parƟcular.21 China, India and the ASEAN countries, see notable increases in gas consumpƟon of around 4-5% in 2035, compared with the New Policies Scenario. In the rest of the world (excluding North America), the increase in 2035 gas use is closer to 3%. In North America, where gas prices are higher than in the New Policies Scenario, demand does not change. This is a result of two oīseƫng trends. On the one hand, gas demand in power generaƟon and Įnal consumpƟon declines somewhat, as prices are higher. On the other, increased producƟon and export push up gas consumpƟon for liquefacƟon and own uses within the oil and gas industry.

1 2 3 4 5 6 7 8

Figure 3.12 ‫ ٲ‬Differences in gas consumption between the Gas Price

bcm

Convergence Case and the New Policies Scenario 120

Rest of world

100

Asia Pacific Europe

9 10

North America

80

11

60

12

40 20

13

0 -20

2025

14

2035

© OECD/IEA, 2013

The inŇux of compeƟƟvely-priced gas from North America implies some displacement of gas producƟon projects elsewhere at the higher end of the internaƟonal cost curve͖ but, 21.ഩ Policies supporting renewables and nuclear do not change in the Gas Price Convergence Case, compared with the New Policies Scenario. However, in regions where natural gas prices are lower and feed-in tariffs are in place for renewables, support schemes become more expensive, creating an extra burden on household and industry bills. This could, in practice, bring the support schemes into question, increasing gas and coal use as a result (although this is not part of the Case considered here). Conversely, in countries with higher gas prices (in North America and parts of Latin America), subsidies to renewables would be lower than in the New Policies Scenario. Overall CO 2 emissions in the Gas Price Convergence Case are very close to those of the New Policies Scenario: lower emissions from coal are largely offset by higher emissions from gas. However, there would be larger reduction in local air pollutants in China, India and ASEAN countries.

Chapter 3 | Natural gas market outlook

135

15 16 17 18

in our modelling, an import price level above $12ͬMBtu in the Asia-PaciĮc region remains suĸcient to bring on addiƟonal producƟon (beyond the levels seen in the New Policies Scenario) from a range of suppliers in various locaƟons, including East Africa, Russian east Siberia and – once pressures on costs have eased – from Australia. The export earnings associated with supply projects are, though, lower than in the New Policies Scenario. While the volume of inter-regional trade in the Gas Price Convergence Case rises to 1ථ150ථbcm by 2035, 5% above the New Policies Scenario, the total value of this trade is 3% lower. Figure 3.13 ‫ ٲ‬Change in import volumes and imports bills for selected

bcm

regions in the Gas Price Convergence Case, relative to the New Policies Scenario, 2035 30

Addional import volume

25

Import bill savings (right axis)

20 15 10

0

0 -5 Japan

Korea

China

India

European Union

-10

Billion dollars (2012)

5

In the Gas Price Convergence Case, gas-imporƟng countries beneĮt from access to lowerpriced gas than in the New Policies Scenario: even for higher volumes of gas, import bills are reduced for all the major gas-imporƟng countries and regions (Figure 3.13). But even though a key aƩracƟon to buyers, access to ͞cheaper gas͟ should not be considered a reliable outcome of this Case – not least because future oil and gas price movements are so uncertain. Indeed, there are no guarantees that gas priced under a system of gas-togas compeƟƟon would always be cheaper than equivalent oil-indexed volumes (even if, at crude oil prices above $100ͬbarrel, this seems likely aside from periods of very cold weather or supply disrupƟon). A more signiĮcant feature of this Gas Price Convergence Case is that regional and global price signals would emerge reŇecƟng the supply-demand fundamentals of the fuel itself, thereby changing the basis of decision-making on new upstream and infrastructure spending and driving gas prices worldwide closer to costs.

© OECD/IEA, 2013

Investment The projected trends in gas demand in the New Policies Scenario imply a need for cumulaƟve investment along the gas-supply chain of almost $8ථ500 billion dollars (in year2012 dollars) over 2012-2035, or around $370 billion per year. Two-thirds of that spending, or $250 billion per year, is needed in the upstream, for new greenĮeld projects and to

136

World Energy Outlook 2013 | Global Energy Trends

combat decline at exisƟng Įelds.22 Transmission and distribuƟon networks account for 23% and LNG faciliƟes – liquefacƟon plants, carriers and re-gasiĮcaƟon terminals – for the remaining 9%. Figure 3.14 ‫ ٲ‬Cumulative investment in natural gas supply infrastructure by region in the New Policies Scenario

LNG

Indonesia Rest of ASEAN OECD Asia Oceania Middle East

Upstream – convenonal

China OECD Europe Africa

6 7

E.Europe/Eurasia OECD Americas 500

1 000

1 500

8

2 000 2 500 Billion dollars (2012)

Well over half of the investment is needed in non-OECD countries, where local demand and producƟon grow the most, though unit costs there tend to be lower than in OECD countries (Figure 3.14). The United States and Canada, where output is projected to rise signiĮcantly, account for one-quarter of total gas investment worldwide͖ investment needs there are boosted by the high capital intensity of unconvenƟonal gas drilling. Projected global annual investment needs to 2035 in the gas industry have risen substanƟally over the last few years, as unit capital costs in both the upstream and downstream sectors have increased, due to increases in the prices of labour, equipment and raw materials. LNG plant costs have also increased rapidly.

© OECD/IEA, 2013

4 5

Upstream – unconvenonal

Lan America

2 3

Transmission and distribuon

India Other developing Asia

1

Although the broad global distribuƟon of gas resources, convenƟonal and unconvenƟonal, gives some cause for comfort, it is far from certain that all the investment required to meet the projecƟons in the New Policies Scenario will be forthcoming. The uncertainƟes apply, in parƟcular, to large upstream and transportaƟon projects (pipeline and LNG), during Ɵmes of transiƟon away from the tradiƟonal models of project Įnancing, based on long-term oil-indexed contracts. Although the gas industry and Įnancial sector have shown their ability to think creaƟvely about ways to miƟgate risks and meet Įnancing requirements, there is a risk that some of the required projects may be delayed. A range of regulatory issues, including domesƟc under-pricing of oil and gas, can also serve as impediments to investment. The experience of Trinidad and Tobago is indicaƟve of the opportuniƟes and piƞalls that can arise as countries seek to capture full economic beneĮts from the development of their hydrocarbon resources (Box 3.8).

9 10 11 12 13 14 15 16 17 18

22.ഩ A more detailed discussion of upstream oil and gas investment and costs can be found in Chapter 14.

Chapter 3 | Natural gas market outlook

137

Box 3.8 ‫ ٲ‬Trinidad and Tobago: seeking a new foothold in a changing gas world Natural gas is instrumental to Trinidad and Tobago’s economy, but a number of challenges will have to be dealt with if it is to remain an important source of revenue for the country. The chosen strategy for making best use of the gas resources has been two-pronged: export both LNG and chemicals. The Įrst LNG exports were made in 1999, and Trinidad and Tobago has since become one of the world’s largest LNG exporters: 14ථmillion tonnes (19ථbcm) were shipped to various countries in 2011. The energy sector, including petroleum, accounts for some 45% of GDP. Large-scale investment in the domesƟc petrochemical industry has been aƩracted by making gas available below the internaƟonal price, turning the country into the world’s largest exporter of ammonia and methanol. But conƟnued growth in LNG and petrochemical exports may be constrained by faltering upstream investment, rising domesƟc consumpƟon and growing compeƟƟon from other exporters. Under-pricing of natural gas sold to the power sector, industry and households, has contributed to a near doubling in gas use between 2000 and 2012. Discoveries have failed to keep pace with the growth of producƟon and proven reserves have dropped by one-third since 2002, to around 375ථbcm. The reserves-toproducƟon raƟo has declined to ten years. In an eīort to regain the dynamism of the past, new incenƟves for upstream investment (primarily tax-related) have led to an increase in exploraƟon acƟvity and the annual decline rate of reserves has been curtailed from 9.5% (2008) to 1.1% (2012). A deepwater licensing round in 2012 awarded four of six blocks on oīer.

© OECD/IEA, 2013

Trinidad and Tobago’s oil sector faces similar challenges. Prior to the dramaƟc increase in natural gas producƟon and exports, oil dominated the economy. ProducƟon peaked at 230ථthousand barrels per day (kbͬd) in the late 1970s and has since been in decline, standing at 120ථkbͬd in 2012. The country remains a net exporter of oil, but levels have been falling, while demand conƟnues to grow, abeƩed by economic growth and subsidised motor fuels. We esƟmate that oil subsidies alone amounted to $290ථmillion in 2012, with the bulk going to transport fuels, limiƟng (directly or indirectly) the resources available to government for other spending prioriƟes and leading to problems like fuel smuggling, which the government is now trying to stamp out, e.g. by means of increased penalƟes. Steps are also being taken to tackle the subsidy issue: the price of premium gasoline was sharply increased in October 2012, and the government is encouraging motorists to switch from oil-based fuels to compressed natural gas by providing Įscal incenƟves and developing re-fuelling infrastructure at service staƟons.

138

World Energy Outlook 2013 | Global Energy Trends

Chapter 4 Coal market outlook Blockbuster or losing lustre? Highlights

x The magnitude of future global coal demand growth is uncertain, parƟcularly because of the varying stringency of environmental policies assumed in the three scenarios. In the New Policies Scenario, global coal demand grows by 0.7% per year over 2011-2035, slowing noƟceably aŌer 2020 as announced policies to foster renewables, penalise CO2 emissions and address other environmental issues take eīect. Demand increases more than twice as fast in the Current Policies Scenario, while in the 450ථScenario coal demand drops by one-third relaƟve to 2011.

x Coal demand trends diverge across regions. In the New Policies Scenario, OECD coal use falls by one-quarter by 2035 as coal is backed out of power generaƟon. By contrast, demand expands by one-third in non-OECD countries – predominantly in India, China and the ASEAN region – despite China reaching a plateau aŌer 2025.

x Globally, coal remains the leading source of electricity generaƟon in the New Policies Scenario, though its share falls from 41% to 33% in 2035. The power sector accounts for around 63% of total coal use in the period to 2035. Industry coal demand growth saturates over the period, however, feedstock in the petrochemicals industry and coal-to-liquids emerge as signiĮcant growth sectors.

x Growth in coal producƟon over 2011-2035 comes mainly from non-OECD countries, with India, Indonesia and China accounƟng for 90% of incremental coal output. Australia is the principal OECD country with higher producƟon. Coal resources will not be a constraint for many decades, yet the cost of supply is likely to increase moderately in real terms as a result of rising mining and transportaƟon costs.

x Already the world’s largest coal user, producer and now importer, China conƟnues to dominate coal markets in the New Policies Scenario. Nonetheless, China’s rate of growth in coal demand is set to slow as eĸciency measures bear fruit, the power sector diversiĮes and industrial coal growth saturates with its peaking steel and cement output. Subject to price arbitrage, China’s coal net imports peak by 2020.

x India becomes the second-largest coal user in the next decade, surpassing the United States. Despite abundant coal resources, domesƟc supply is not keeping pace with demand, which has caused imports to double since 2008. India overtakes Japan and the European Union within a few years and China early in the next decade to become the world’s largest coal importer, with imports reaching 350ථMtce in 2035. © OECD/IEA, 2013

x The eĸciency of China’s coal plants is improving. However, many units under construcƟon or planned in the ASEAN region and India use subcriƟcal technologies, which consume up to 15% more coal for a given power output than more eĸcient supercriƟcal technologies and lock in higher CO2 emissions for decades to come. Chapter 4 | Coal market outlook

139

Overview Coal use has increased substanƟally over the last decade, driven by and contribuƟng to an unprecedented expansion in economic acƟvity, as well as reducing poverty across the developing world (Figureථ4.1). In fact, coal provided nearly half of the increase in global primary energy demand over the decade to 2012 (Boxථ4.2).1 But its use has serious drawbacks, especially if ineĸcient: coal is a major source of local air polluƟon and, as the most carbon-intensive fossil fuel, it is the main contributor to rising energy-related carbon dioxide (CO2) emissions. The magnitude of future coal demand growth hinges criƟcally on the acƟons that governments take to address these issues, taking into account their aspiraƟons for energy security, aīordability and improved access to modern energy.2 The wide divergence in outcomes for coal in the three scenarios, notably aŌer 2020, reŇects primarily the diīerent degrees of stringency of the policies adopted to promote energy eĸciency, reduce greenhouse-gas emissions and improve local air quality (see AnnexථB). Put another way, the diīerences in the scenarios reŇect the importance of coal use – in parƟcular in the power sector – as a factor in energy and climate change policies. Figure 4.1 ‫ ٲ‬Incremental world coal demand, historical and by scenario 1987-2001 1987-2011

2001-2011 2011-2020

New Policies Scenario

2020-2035

Current Policies Scenario 450 Scenario -2 000

-1 000

0

1 000

2 000

3 000 Mtce

© OECD/IEA, 2013

In the New Policies Scenario, which assumes the cauƟous implementaƟon of announced policy measures, growth in global coal demand averages 0.7% per year over 2011-2035. This is a marked slowdown compared with the 2.5% averaged over the past 25ථyears. Coal demand expands from around 5ථ390ථmillion tonnes of coal equivalent3 (Mtce) in 2011 (5ථ425ථMtce in 2012 based on preliminary data) to 6ථ325ථMtce in 2035. Two-thirds of this occurs in the period to 2020, with demand growing by only 0.4% per year thereaŌer. 1.ഩ For 2012, preliminary data for aggregate coal demand, production and trade by country are available͖ the sectoral breakdown for coal demand is estimated (complete data are available to 2011). 2.ഩ It is estimated that around 1.3ථbillion people (18% of the world’s population) did not have access to electricity in 2011͖ around 2.6ථbillion (38% of the world’s population) relied on the traditional use of biomass for cooking (see Chapterථ2). 3.ഩ A tonne of coal equivalent equals 7ථmillion kilocalories (kcal) or equivalent to 0.7ථtonnes of oil equivalent. 140

World Energy Outlook 2013 | Global Energy Trends

Demand increases strongly in non-OECD countries, more than oīseƫng a decline in the OECD (Tableථ4.1). Nearly three-quarters of the increase in global coal demand comes from the power sector, even though coal’s share of global electricity generaƟon declines by eight percentage points as many countries diversify their power mixes. At 33%, coal remains the leading source of electricity generaƟon in 2035. Coal producƟon today is dominated by non-OECD countries, whose share of output rises further in all three scenarios.

1

Table 4.1 ‫ ٲ‬Coal demand, production and trade by scenario (Mtce)

4

New Policies

OECD Non-OECD

World

CurrentථPolicies

450 Scenario

1990

2011

2020

2035

2020

2035

2020

2035

Demand

1 543

1 518

1 469

1 156

1 524

1 502

1 264

627

ProducƟon

1 533

1 397

1 430

1 300

1 536

1 697

1 215

691

Demand

1 643

3 872

4 533

5 170

4 880

6 262

4 043

2 992

ProducƟon

1 661

4 101

4 573

5 026

4 868

6 066

4 092

2 928

Demand

3 186

5 391

6 003

6 326

6 404

7 764

5 307

3 619

Steam coal

2 244

4 220

4 689

5 152

5 049

6 440

4 067

2 712

Coking coal

542

858

993

929

1 025

1 017

959

810

Lignite

400

313

321

246

330

307

281

97

3 194

5 498

6 003

6 326

6 404

7 764

5 307

3 619

309

900

1 152

1 261

1 295

1 649

958

635

ProducƟon Inter-regionalථtrade* Steam coal

162

652

850

922

975

1 276

672

383

Coking coal

186

255

316

348

331

388

300

267

* Total net exports for all WEO regions, not including trade within regions. Notes: Historical data for world demand diīer from world producƟon due to stock changes. Lignite also includes peat.

© OECD/IEA, 2013

Restrained less by climate change policy intervenƟon, coal demand in the Current Policies Scenario grows more than twice as fast as in the New Policies Scenario. The increase of around 2ථ375ථMtce is slightly less than over the last 25ථyears. In OECD countries, coal demand in the Current Policies Scenario falls only marginally by 2035, unlike the marked decline in the New Policies Scenario. The strong growth in coal demand in non-OECD countries in the Current Policies Scenario results in global coal use overtaking oil use soon aŌer 2020 and coal remains the leading fuel throughout the period to 2035. The 450 Scenario, which assumes that strong policy measures are implemented to keep long-term greenhouse-gas-induced temperature changes to 2ථdegreesථCelsius, sees global coal use fall by 33% over 2011-2035. This is a return to the level of demand in the early 2000s. As a result, coal’s share in the global energy mix declines by twelveථpercentage points, reaching 17% in 2035. Coal demand in the power sector is cut by more than half during the projecƟon period, with the fuel providing only 14% of global electricity generaƟon in 2035, compared with 33% in the New Policies Scenario. Of total coal-Įred electricity output in 2035, nearly 60% comes from plants ĮƩed with carbon capture and storage (CCS) technology.

Chapter 4 | Coal market outlook

141

2 3

5 6 7 8 9 10 11 12 13 14 15 16 17 18

Demand for steam coal varies more than that for coking coal across the three scenarios, since steam coal is used mainly (70%) for power generaƟon – the sector that is most aīected by local air polluƟon and climate change policies. In the New Policies Scenario, steam coal use in 2035 is four-ĮŌhs of the level projected in the Current Policies Scenario, but nearly twice that in the 450ථScenario. InternaƟonal steam coal trade is more strongly aīected: relaƟve to demand, small volumes (15%) of steam coal are traded and, consequently, small changes in demand or supply can impact trade disproporƟonately. In the Current Policies Scenario, steam coal trade between the main WEO regions increases by half in the period to 2020 and conƟnues to rise steadily thereaŌer, with trade nearly doubling over the enƟre projecƟon period. In the New Policies Scenario, by contrast, steam coal trade grows by 30% over 2011-2020, but slows thereaŌer. In the 450ථScenario, steam coal trade peaks at about 765ථMtce around 2015 and then falls steeply to half that level by 2035. Coking coal is less easily subsƟtuted in industrial applicaƟons and so demand and trade are far less aīected by government policies (Boxථ4.1). In all three scenarios, coking coal trade underpins a signiĮcant share (around 35%) of global coking coal use in 2035. Even in the 450ථScenario, coking coal trade in 2035 remains at around 2011 levels. Box 4.1 ‫ ٲ‬A quick guide to the different types of coal4 Coal is a generic name given to a wide range of solid organic fuels of varying composition (e.g. volatile matter, moisture, ash and sulphur content or other impurities) and energy content. For convenience, the IEA divides coal into three distinct categories: „Steam coal accounts for nearly 80% of global coal demand today. It is mainly used

for heat producƟon or steam-raising in power plants (70%) and, to a lesser extent, in industry (15%). Typically, steam coal is not of suĸcient quality for steel making. „Coking coal accounts for around 15% of global coal demand. Its composiƟon makes

it suitable for steel making (as a chemical reductant and source of heat), where it produces coke capable of supporƟng a blast furnace charge. „Lignite accounts for 5% of global coal demand. Its low energy content and usually

© OECD/IEA, 2013

high moisture levels generally make long-distance transport uneconomic. Over 90% of global lignite use today is in the power sector. Data on lignite in the WEO includes peat, a solid formed from the parƟal decomposiƟon of dead vegetaƟon under condiƟons of high humidity and limited air access. The rapid build-up of new coal-Įred power staƟons, many using relaƟvely ineĸcient subcriƟcal technology, runs the risk of a large-scale lock-in of CO2 emissions for decades ahead, notably in non-OECD countries. One possible set of measures to address climate change is to ensure that ineĸcient subcriƟcal power plants are no longer built, and to limit the use of exisƟng ones where possible without puƫng the reliability of electricity 4.ഩ Detailed classifications and definitions of coal types are available in WEO-2011 (IEA, 2011). 142

World Energy Outlook 2013 | Global Energy Trends

supply at risk (IEA, 2013a). Beyond 2020, when demonstrated and deployed at new high eĸciency plants, or retroĮƩed at suitable exisƟng plants, CCS may play a key role in curbing CO2 emissions from coal-based power generaƟon and industry (see Chapterථ1). As such, CCS could act as an asset protecƟon strategy, enabling more fossil fuels to be used and potenƟally reducing the overall cost of power sector decarbonisaƟon by around $1ථtrillion between 2012 and 2035 (IEA, 2013a).

1

Demand

4

Regional trends In the New Policies Scenario, coal demand trends conƟnue to diverge across regions (Figureථ4.2). Coal demand in OECD countries declines further throughout the projecƟon period, with the fall acceleraƟng aŌer 2020 as renewables and gas increase their combined share of electricity generaƟon in Europe, the United States and Asia Oceania. In Europe, coal use falls steeply: in 2035 it is just 57% of its 2011 level and accounts for only 11% of OECD Europe’s electricity needs in 2035, compared with 25% today. The role of coal is reduced by the growth of renewables and the reƟrement of old coal-Įred plants at a faster rate than new coal capacity is commissioned (see Chapterථ5). In the United States, coal demand falls at a more modest pace, with the decline acceleraƟng aŌer 2020, as the reƟred old coal-Įred plants are replaced by renewables and gas. In Japan, coal for power generaƟon also slides over the Outlook period as renewables gain market share.

5 6 7 8 9

11

Rest of world South Africa

80%

3

10

Figure 4.2 ‫ ٲ‬Coal demand by key region in the New Policies Scenario 100%

2

12

Russia European Union

60%

Japan and Korea

13

United States 40%

ASEAN

14

India 20%

China

© OECD/IEA, 2013

1990

2000

2005

2011

2020

15

2035

16

Coal demand in non-OECD countries conƟnues to increase in the New Policies Scenario, though at a much slower rate than in the last two decades. Rates of growth vary widely across non-OECD countries (Tableථ4.2). China and India, which possess large, relaƟvely lowcost indigenous resources, remain the main centres of coal use, with their combined share of global demand rising from 58% in 2011 to 64% in 2035. zet trends diīer markedly in the Chapter 4 | Coal market outlook

143

17 18

Box 4.2 ‫ ٲ‬Was 2012 an aberration or a harbinger of change in coal demand? Coal use conƟnued to grow strongly in 2012 in several large coal-consuming countries, according to preliminary data. In Japan and India, it grew by around 6% and 4%. Coal is a major source of electricity generaƟon in Japan (nearly 30%) and is set to grow in the short term, especially in the context of reduced nuclear output aŌer the Fukushima Daiichi accident. Some European countries, where gas was relaƟvely expensive and CO2 prices were low (see Boxථ5.1 in Chapterථ5), also registered substanƟal increases in coal use, notably the Unitedථ
© OECD/IEA, 2013

The answer lies, perhaps, somewhere in between the two. In China, a slowdown in the rate of economic growth is leading to a marked deceleraƟon in energy demand growth, and coal demand growth in parƟcular. Coal use in industry, which makes up nearly onequarter of the country’s total coal demand, increased by 4.5% in 2012, compared with 7.5% on average in the last decade. Demand in the power sector, which accounts for over half of China’s coal demand, rose marginally, compared with double-digit growth rates in prior years. To some extent, this was the result of China adding a record 16ථgigawaƩs (GW) of hydro capacity and 2012 being a parƟcularly wet year, factors that boosted hydro output and reduced the need to run baseload coal plants. But the slowdown in coal demand growth also reŇects progress by China in promoƟng energy eĸciency and diversifying its power sector. Projects due to come online in the period to 2020 are expected to further limit the need for increased coal consumpƟon (Figureථ4.7). In the United States, a very diīerent phenomenon reduced coal use in the power sector (which accounts for 90% of US coal use) in 2012. The rise of unconvenƟonal gas producƟon, coupled with a historically mild winter, saw Henry Hub gas prices fall to as low as $1.82 per million BriƟsh thermal units (MBtu) in April 2012, making gas highly compeƟƟve against coal in the sector, especially in eastern regions where coal is relaƟvely expensive. This led to gas displacing coal on a large scale. By early September 2013, gas prices rebounded to an average of $3.7ͬMBtu for the year to date, leading to coal regaining market share. In the Įrst half of 2013, coal-Įred plants accounted for just under 40% of electricity output, compared with an average of 35% in the Įrst half of 2012 (US EIA, 2013). Nonetheless, in the longer term, renewables and gas are expected to gradually displace coal for electricity generaƟon as US environmental restricƟons on coal burning become more stringent (see Boxථ4.4).

144

World Energy Outlook 2013 | Global Energy Trends

two countries, reŇecƟng in large part their diīerent stages of economic development. In China, growth in coal use slows over 2020-2030 and stabilises around the end of the projecƟon period as a result of lower electricity demand growth and other fuels gaining market share (Spotlight). Therefore China’s share of global coal demand, which has risen notably in the past decade, levels oī at around half. In India, coal use conƟnues to grow briskly throughout the projecƟon period, in line with the country’s strong electricity demand growth. India displaces the United States as the world’s second-largest coal market before 2025. AssociaƟon of Southeast Asian NaƟons (ASEAN) countries see a tripling of coal use͖ their collecƟve consumpƟon is nearly double that of the European Union in 2035.

1

Table 4.2 ‫ ٲ‬Coal demand by region in the New Policies Scenario (Mtce)

5

OECD Americas United States

1990

2011

2020

2025

2030

2035

1ථ543

1 518

1ථ469

1ථ369

1ථ262

701

734

714

683

657

684

657

625

2011-2035 Delta

CAAGR*

1ථ156

-362

-1.1%

649

631

-103

-0.6%

598

587

-97

-0.6%

Europe

645

445

408

351

302

253

-193

-2.3%

Asia Oceania

198

339

347

335

311

272

-67

-0.9%

109

153

157

153

149

140

-13

-0.4%

1 643

3 872

4ථ533

4ථ792

4ථ993

5ථ170

1 298

1.2%

525

329

334

337

338

346

17

0.2%

Japan Non-OECD E. EuropeͬEurasia Russia

273

166

165

170

171

175

9

0.2%

991

3 355

3 974

4 211

4 403

4ථ561

1 206

1.3%

China

762

2 666

3 026

3 094

3 095

3ථ050

384

0.6%

India

148

465

607

713

840

972

507

3.1%

18

129

224

275

331

399

269

4.8%

1

4

6

7

7

8

4

2.7%

Asia

ASEAN Middle East

106

152

176

185

188

194

41

1.0%

South Africa

95

140

151

155

153

154

14

0.4%

LaƟn America

21

32

44

52

58

61

30

2.8%

Africa

Brazil World European Union

14

22

27

30

32

34

12

1.9%

3ථ186

5ථ391

6ථ003

6ථ160

6ථ255

6ථ326

936

0.7%

651

409

356

300

250

207

-202

-2.8%

4

6

8 9 10 11 12 13 14 15 16

Sectoral trends

© OECD/IEA, 2013

3

7

* Compound average annual growth rate.

The power sector has been an increasingly dominant source of global coal demand. Over 1990-2011, its share of coal use rose from 55% to 63%. This trend slows considerably in the New Policies Scenario, as the sector’s share increases only marginally between 2011 and 2035. This reŇects a rapid decline in coal use in the power sectors of OECD countries being oīset by conƟnued growth in non-OECD countries (Figureථ4.3). Chapter 4 | Coal market outlook

2

145

17 18

S P O T L I G H T Is China’s coal demand set to peak soon? In the decade to 2011, China’s coal use more than doubled and its share in global coal demand rose from 30% to nearly 50%. There is liƩle doubt that such strong growth will taper oī in the future, with some industry analysts even expecƟng China’s coal demand to peak in the current decade. In 2012, the rate of coal demand growth in China was one of the lowest over the past decade: the drivers of change may already be at work. China’s 12th Five-zear Plan (the Plan), adopted in 2011, includes targets to reduce energy intensity by 16% and cut CO2 intensity by 17% by 2015, compared with 2010. Measures to achieve overarching targets have been reinforced by detailed industry targets, parƟcularly in terms of diversifying the power sector, which today accounts for some 55% of China’s coal use. The diversiĮcaƟon aims to add 70ථgigawaƩs (GW) of wind and 35ථGW of solar capacity by 2015, and to start construcƟon by that year of 120ථGW of hydro and 40 GW of nuclear capacity. In addiƟon, growing public concerns about deterioraƟng air quality in many ciƟes have led Chinese authoriƟes to announce further measures to reduce polluƟon from parƟculate maƩer, including greater use of natural gas. Will these measures curb coal demand growth? zes, but to reduce coal demand, the rate of new technology deployment and eĸciency improvements would have to outpace power demand growth. To reduce air polluƟon, the likely near-term soluƟon will be to shut down some of the dirƟest power plants and steel mills while installing or enforcing the use of emissions control technologies at newer plants. In addiƟon to coal, oil use in transport impacts air quality and has a key role in this debate.

© OECD/IEA, 2013

The Plan also envisages rebalancing away from energy-intensive industry and stronger pursuit of energy eĸciency gains, while moving away from double-digit GDP growth. This will curb China’s coal demand growth, yet its household electricity consumpƟon today is signiĮcantly lower than the OECD average. How quickly the economy can be rebalanced depends on global demand for Chinese products. Bold policy measures, an economic ͞hard landing͟ or a rapid shiŌ to a services-based economy are potenƟal game-changers that could unexpectedly impact China’s coal demand. ShiŌing away from a fuel that presently accounts for almost 70% of China’s total energy demand and 80% of its electricity output is expected to take Ɵme. In the New Policies Scenario, China’s coal demand growth slows before 2020, with demand reaching a plateau aŌer 2025. Even then, coal conƟnues to dominate China’s total primary energy demand (57%) and electricity generaƟon (59%). Costs are on the rise in mature mining regions in northern China and long transport distances to the demand hubs can make some domesƟc coal expensive. Low-cost suppliers from the internaƟonal market will therefore remain compeƟƟve against high-cost domesƟc producers, especially in China’s southern coastal provinces. Hence, in our projecƟons, China conƟnues to import substanƟal amounts of coal, remaining a strong force in global coal markets.

146

World Energy Outlook 2013 | Global Energy Trends

Industry, including own use and transformaƟon in blast furnaces and coke ovens, accounts for most of the remaining global coal demand. Coal use in industry conƟnues to expand rapidly in the New Policies Scenario unƟl 2020, and then begins to decline. AŌer 2020, other energy sources and energy eĸciency measures are deployed more widely in nonOECD countries (mirroring past trends in the OECD) and global crude steel output ŇaƩens (around 2030). Overall, the share of industrial energy supplied by coal falls globally from 27% in 2011 to 24% in 2035. Iron and steel producƟon remains the largest coal user in industry. Coal use in industry peaks around 2020 as other technologies (such as electric arc furnaces) become more widespread, eĸciency improvements are achieved and crude steel producƟon in China begins to decline. There is also a signiĮcant increase in the use of coal as a feedstock in the petrochemicals industry and in coal-to-liquids plants, notably in China (seeථChapterථ15).

OECD

4 5

245 Mtce

8

Industry* 2035

3

7

Power generaon 1 218 Mtce

2

6

Figure 4.3 ‫ ٲ‬Coal demand by key sector in the New Policies Scenario

2011

1

864 Mtce

Coal-to-liquids and petrochemical feedstocks

210 Mtce

9

Non-OECD

Buildings 2011

2 160 Mtce

1 192 Mtce

10

Other 2035

3 202 Mtce 20%

40%

1 325 Mtce 60%

80%

11 100%

12

* Includes own use and transformaƟon in blast furnaces and coke ovens.

Supply

13

Resources and reserves5

© OECD/IEA, 2013

Coal is the most abundantly available fossil fuel worldwide (despite large recent addiƟons to natural gas resources), and the resource base is easily suĸcient to meet any plausible level of demand for decades to come. Proven coal reserves – volumes that are known to exist and are thought to be economically and technically exploitable at today’s prices – totalled around 1ථ040ථbillion tonnes at the end of 2011, of which coking and steam coal make up nearly three-quarters (BGR,ථ2012).6 Proven coal reserves worldwide have increased by around 35ථbillion tonnes in 2011 (producƟon was around 7.7ථbillion tonnes) thanks to reserve addiƟons mainly in Australia, South Africa and Indonesia. Coal consƟtutes around 5.ഩථA detailed assessment of coal resources and production technologies is available in WEO-2011 (IEA, 2011). 6.ഩථClassifications of coal types can differ between BGR and IEA due to statistical allocation methodologies.

Chapter 4 | Coal market outlook

147

14 15 16 17 18

55% of the world’s total proven fossil fuel reserves. Resources are more than twenty Ɵmes larger than reserves and make up about 90% of global remaining fossil fuel resources. A signiĮcant porƟon can be readily exploited with relaƟvely modest price increases or technical innovaƟons. Coal reserves and resources are widely distributed: 32ථcountries have reserves of more than 1ථbillion tonnes (roughly the level of annual internaƟonal coal trade)͖ 26ථcountries have resources of more than 10ථbillion tonnes (BGR, 2012). Non-OECD Asia contains around 30% of the world’s proven reserves and Australia has a further 10%. In all major producer countries, proven reserves comfortably exceed their projected cumulaƟve producƟon to 2035 in the New Policies Scenario (Figureථ4.4). In several of these countries, cumulaƟve producƟon over 2011-2035 could in theory be met by drawing solely on the reserves of mines currently operaƟng, but in pracƟce this would be impeded by capacity constraints. Figure 4.4 ‫ ٲ‬Reserves and cumulative production by major coking and steam coal producers in the New Policies Scenario Cumulave producon:

United States

2012-2035

China

1800-2011

India

Reserves end-2011:

Russia

Operang mines

Australia

Remaining

South Africa Indonesia -100

-50

0

50

100

150

200 250 Billion tonnes

Sources: Etemad, et al. (1991)͖ BGR (2012)͖ Wood Mackenzie databases͖ IEA analysis.

WƌŽĚƵĐƟŽŶ

© OECD/IEA, 2013

While coal resources will not constrain coal producƟon for many decades, supply costs are likely to increase in real terms as a result of rising input costs, such as fuel or explosives (the cost of which is oŌen closely linked to oil prices), and higher labour costs. The development of mines in more remote or undeveloped regions, where infrastructure and transport costs tend to be higher, will also add to coal supply costs. Steam coal prices rose sharply over the period 2009-2011, outpacing rising producƟon costs and incenƟvising investments in mining, processing and transportaƟon faciliƟes (Figureථ4.6). Since 2011, however, steam coal prices have fallen while costs have conƟnued to climb (and even accelerate in some countries), weakening investment incenƟves. The prospect of conƟnued low coal prices and cost pressures (in the absence of strong producƟvity gains) may limit investment and producƟon growth in some countries. AddiƟonally, potenƟal export bans and policy intervenƟons related to environmental policies can signiĮcantly

148

World Energy Outlook 2013 | Global Energy Trends

impact coal demand and investment. In the New Policies Scenario, which assumes cauƟous implementaƟon of climate change policies and greater compeƟƟon between natural gas and coal in the power sector, the average OECD steam coal import price rises from $99ථper tonne (inථyear-2012 dollars) in 2012 to $106ͬtonne in 2020 and then more slowly to $110ͬtonne in 2035 (seeථChapterථ1). Coal producƟon prospects diīer markedly between OECD and non-OECD countries (Tableථ4.3). In the New Policies Scenario, OECD coal output grows modestly to 2020, falling steadily thereaŌer. OECD Europe experiences a halving of coal output, reŇecƟng the phase-out of coal mining subsidies in some countries as well as cost escalaƟon in coal Įelds that have been producing for many decades. The United States, by far the largest coal producer among OECD countries, sees a slower decline in producƟon of 0.7% per year over 2011-2035. Although some mature coal basins in the United States are experiencing cost escalaƟon, there are sƟll large quanƟƟes of coal that can be extracted at relaƟvely low costs (Figureථ4.10). Australia’s producƟon grows by almost 50% between 2011 and 2035, fuelled by rising exports. Table 4.3 ‫ ٲ‬Coal production by region in the New Policies Scenario (Mtce) 2020

2025

2030

2035

1 533

1 397

1 430

1 384

1 343

1 300

-97

-0.3%

836

826

797

768

728

700

-126

-0.7%

775

766

735

708

674

653

-113

-0.7%

Europe

526

248

218

180

151

123

-125

-2.9%

Asia Oceania

171

323

415

435

464

478

154

1.6%

Australia

152

318

410

431

459

473

155

1.7%

Non-OECD

1 661

4 101

4 573

4 776

4 912

5 026

925

0.9%

533

429

448

437

433

432

3

0.0%

275

257

269

264

261

258

1

0.0%

United States

E. EuropeͬEurasia Russia Asia

CAAGR*

952

3 377

3 755

3 945

4 069

4 162

785

0.9%

741

2 605

2 779

2 860

2 871

2 835

230

0.4%

India

150

360

392

451

527

624

263

2.3%

8

296

449

489

519

549

254

2.6%

Middle East Africa

1

1

1

1

1

1

0

1.1%

150

209

244

259

264

277

68

1.2%

South Africa

143

204

224

231

229

231

28

0.5%

LaƟn America

25

85

125

134

146

155

70

2.5%

Colombia

20

80

116

124

134

142

62

2.4%

3 194

5 498

6 003

6 160

6 255

6 326

829

0.6%

528

239

202

162

130

103

-136

-3.5%

World European Union © OECD/IEA, 2013

Delta

China Indonesia

* Compound average annual growth rate. Note: Historical data and the global CAAGR diīer from world demand in Tableථ4.2 due to stock changes.

Chapter 4 | Coal market outlook

3 4 5 6 7 8

2011

Americas

2

2011-2035

1990 OECD

1

149

9 10 11 12 13 14 15 16 17 18

The picture diīers outside of OECD countries. Output in China conƟnues to increase in line with domesƟc demand, but at a much slower rate than in recent years, with most growth coming before 2020. ProducƟon growth saturates before 2030, at a level 280ථMtce higher than in 2011. Indonesia and India both increase their coal producƟon rapidly in response to growing domesƟc demand and, in the case of Indonesia, to meet growing exports. Colombia and South Africa also expand output, while new producers, such as Mongolia and Mozambique, ramp up their producƟon.

Trade The recent shiŌ in the paƩern of internaƟonal trade, with developing Asian countries relying increasingly on imports, is set to conƟnue in the New Policies Scenario (Tableථ4.4). Responding to global demand trends, coal trade between WEO regions is projected to rise from 900ථMtce in 2011 to 1ථ150ථMtce in 2020. It increases at a more subdued pace aŌer 2020, reaching 1ථ260ථMtce in 2035. Table 4.4 ‫ ٲ‬Inter-regional coal trade in the New Policies Scenario

OECD Americas United States Europe Asia Oceania Australia Japan Non-OECD E.ථEuropeͬEurasia Russia Asia

2020 Share of Mtce demand* -40 3%

2035 Share of Mtce demand* 144 12%

2011-35 Delta Mtce 259

85

11%

82

11%

68

10%

-16

79

11%

77

11%

66

11%

-12

-197

65%

-190

70%

-130

74%

-67

-2

1%

68

17%

206

44%

208 149

263

89%

337

86%

412

90%

-154

100%

-157

100%

-140

100%

-13

157

4%

40

1%

-144

3%

-301

98

28%

115

31%

86

23%

-13

92

42%

104

45%

84

38%

-9

-48

1%

-219

6%

-400

9%

352

China

-129

5%

-247

8%

-215

7%

86

India

-106

24%

-214

37%

-349

37%

242 134

Indonesia

251

85%

363

81%

385

70%

Middle East

-3

75%

-5

81%

-6

83%

3

Africa

56

27%

69

28%

83

30%

27

South Africa

63

31%

73

33%

77

33%

14

LaƟn America

53

63%

81

65%

93

61%

40

Colombia World** European Union © OECD/IEA, 2013

2011 Share of Mtce demand* -115 9%

75

94%

110

94%

134

94%

59

900

17%

1ථ152

20%

1ථ261

21%

361

-170

62%

-155

67%

-105

72%

-66

*ථProducƟon in net-exporƟng regions. **ථTotal net exports for all WEO regions, not including intra-regional trade. Notes: PosiƟve numbers denote net exports and negaƟve numbers denote net imports. The diīerence between OECD and non-OECD in 2011 is due to stock changes. 150

World Energy Outlook 2013 | Global Energy Trends

Asia is set to consolidate its posiƟon as the centre of gravity of internaƟonal coal trade (Boxථ4.3). China became a net importer of coal in 2009: just three years later it was the world’s largest net importer. At about 220ථMtce – a world record – China’s net imports in 2012 were larger than total coal use in any OECD country except the United States. In the New Policies Scenario, China’s net imports of coal peak before 2020, as lower demand growth and improvements in mining producƟvity weaken price diīerenƟals between domesƟc producƟon and imported coal. Nonetheless, imports stay above or around 2012 levels throughout the projecƟon period. The sheer size of China’s coal demand and producƟon means that even small changes in either demand or output can have a very large impact on its import needs. In India, coal imports conƟnue to climb throughout the Outlook period, more than tripling by 2035. India became the world’s fourth-largest importer in 2012, overtaking
© OECD/IEA, 2013

2 3 4 5 6 7 8

While around 85% of global steam coal is sƟll produced and used domesƟcally, internaƟonal trade in steam coal has Ňourished, nearly doubling from 1990 to 2000, and doubling again in the period to 2011. According to preliminary data, worldwide steam coal trade grew strongly in 2012, compared with marginal demand growth, taking the expansion in the trade since 2007 to 50%. Over two-thirds of steam coal trade now serves Asia, both OECD and non-OECD countries. Japan and
Chapter 4 | Coal market outlook

1

151

9 10 11 12 13 14 15 16 17 18

second- and third-largest steam coal exporters. The United States remains an important net exporter of coal through to 2035, with coking coal dominaƟng. Because there are fewer opportuniƟes to subsƟtute for coking coal (contrasƟng with steam coal), major producers and exporters of coking coal – namely Canada, the United States and Australia – are less aīected by the environmental policies of imporƟng countries. Mozambique could emerge as an important new source of export growth, though the Ɵmeline for it to do so remains uncertain. Recently, coal producers’ plans to ship coal via barge on the ambezi River were not approved because of environmental concerns. In addiƟon, heavy rainfall and threats from rebel groups have hampered coal transport on the key railway line recently. With current infrastructure being insuĸcient to accommodate large-scale exports, industry and the government are discussing plans to build a rail link to the coast that would allow up to 25ථmillion tonnes (Mt) per year of coal exports. In the New Policies Scenario, Mozambique’s exports are projected to rise to 6ථMtce in 2020 and nearly 20ථMtce in 2035.

Costs and investment Since the capital costs of coal producƟon are relaƟvely low compared with oil and gas, it is the variable costs of coal supply – oŌen termed the cash costs – that determine the compeƟƟveness of individual exporters. The cash costs of coal exports set the minimum price that an exporter could charge to cover the operaƟng expenses of a mine. However a further margin is required to provide an adequate return on capital expenditure and to aƩract new capital investment. The evoluƟon of coal prices is closely linked to supply cost developments worldwide. The fundamental cost drivers of an exisƟng mine, many of which are likely to remain in producƟon through to 2035, are input costs such as labour, fuel, explosives and maintenance costs. The costs of a new mine are determined chieŇy by the geological condiƟons of the deposit, access to infrastructure and the distance to transport coal to the point of sale.

© OECD/IEA, 2013

Even though some coal exporters need prices in excess of $80ͬtonne (excluding sea freight to the customer) to cover just their cash costs, the bulk of internaƟonally traded steam coal is currently available at a free on board (FOB) cash cost7 of $40-60ͬtonne (Figureථ4.5). This cash cost has shiŌed up signiĮcantly over the last few years. Exchange rates are a major factor. The mechanism is simple: as internaƟonal coal trade is seƩled mostly in US dollars, coal exporters generate a revenue stream in US dollars, while they incur a large part of their costs in local currency. Therefore, a devaluaƟon of the US dollar reduces returns to producers from the internaƟonal market.8 The currencies of most coal-exporƟng countries 7.ഩ FOB (free on board) cash cost includes mining costs, costs of coal washing and preparation, inland transport, mine overhead as well as port charges (this definition corresponds to the C1-cash cost definition widely used in the mining industry). It excludes royalties and taxes, as well as seaborne shipping costs. 8.ഩ Coal importers, buying coal in US dollars and selling, for example, electricity, in local currency, are exposed to a converse effect: a devaluation of the US dollar lowers their procurement cost compared with the value of their product in local currency. 152

World Energy Outlook 2013 | Global Energy Trends

have risen against the dollar in the past three years, especially the Australian dollar (A$). With increasing labour costs, this has meant that Australia has shiŌed, in terms of US dollars, from a mid-cost supplier to being a high-cost supplier within a few years. With increases in other operaƟng costs and construcƟon costs, both exisƟng and new projects are under strain and some new projects are experiencing serious delays. Some $30ථbillion of new coal mining and infrastructure (notably port expansions) in Australia have been delayed or cancelled in the last year (BREE, 2013).

Dollars per tonne (adjusted to 6 000 kcal/kg)

South Africa Colombia

100

Australia

60

United States Russia Other

100

200

300

400

500

600

7

9

700 800 850 Million tonnes

10

Sources: Wood Mackenzie databases and IEA analysis.

© OECD/IEA, 2013

5

8

20

0

3

6

Indonesia 80

40

2

4

Figure 4.5 ‫ ٲ‬FOB cash costs for seaborne steam coal exports, 2012 120

1

Local factors have also contributed to higher cash costs in many cases. Indonesia, the largest steam coal exporter, has built market share rapidly over the past decade on the basis of low mining and transport costs, using barges instead of generally more expensive rail systems to get exports to coastal shipping points. However, the last two years have seen sharp increases in Indonesian operaƟng costs, with higher labour costs and oil prices being major factors, as well as worsening qualiƟes of coal seams in some operaƟons. Nonetheless, Indonesia remains in the lower half of the global steam coal cash cost curve, underpinning its ongoing rapid expansion. Steam coal exports have increased nearly Įvefold over the decade to 2010 and by a further 40% (or 88ථMtce) in the last two years.

11

In the New Policies Scenario, cumulaƟve global investment in the coal supply chain (new coal mines, ports and shipping) amounts to $860ථbillion (inථyear-2012 dollars) over 20132035, of which the majority is in mining. The coal supply chain accounts for only around 2.5% of total investment globally in fuel supply and power generaƟon. This reŇects that coal has a relaƟvely low capital intensity and that It grows much slower than other energy sources, such as renewables and gas, over the Outlook period. Projected coal supplychain investments are centred in non-OECD countries, which account for three-quarters of the total. China alone accounts for 55% of the non-OECD total. Nearly 60% of OECD coal investments, or $90ථbillion, are made in Australia.

15

Chapter 4 | Coal market outlook

153

12 13 14

16 17 18

WƌŝĐŝŶŐŽĨŝŶƚĞƌŶĂƟŽŶĂůůLJƚƌĂĚĞĚĐŽĂů Developments in the power sector are, and will remain, one of the most important demand-side inŇuences on coal prices and trade. Even moderately higher natural gas prices, compared with today’s levels in the United States, would probably trigger a rise in the share of coal in its power supply, potenƟally reducing US steam coal exports. And even a modest rise in global coal prices might see China’s coal imports falter from their current record levels, as uƟliƟes and factories located in coastal China are adept at arbitraging domesƟc and internaƟonal coal prices. By contrast, India’s imports are less sensiƟve to price. Policy decisions in China and India, or even discussions of possible policy changes, will undoubtedly strongly inŇuence internaƟonal coal markets. The global coal market consists of various regional sub-markets, which are typically separated according to geography, coal quality or infrastructure constraints. The internaƟonal market for coal is a comparaƟvely small sub-market (around 17% of global coking and steam coal producƟon is traded internaƟonally), yet it plays a key role as it links the various domesƟc markets through imports, exports and price movements. The degree to which regional coal prices Ňuctuate with price movements on the internaƟonal market depends on how well the regional and internaƟonal markets are connected. US coal prices are a good example: while western coal prices are quite Ňat and low, prices in the eastern United States Ňuctuate strongly with internaƟonal prices as local producers have the opƟon to export. InternaƟonal coal markets have seen sharp price ŇuctuaƟons in recent years, followed by a general price decline in the last two years (Figureථ4.6): from levels of around $120ͬtonne for steam coal in most of 2011, prices have declined to $80-90ͬtonne in 2013.

Index (1Q 05 = 1)

Figure 4.6 ‫ ٲ‬Quarterly indices for IEA crude oil and steam coal prices 3.5 3.0 2.5 IEA crude oil 2.0 Qinhuangdao

1.5

Richards Bay 1.0

Northwest Europe

0.5

© OECD/IEA, 2013

1Q 05

1Q 06

1Q 07

1Q 08

1Q 09

1Q 10

1Q 11

1Q 12

3Q 13

Notes: Qinhuangdao is a major coal port in northeast China. Richards Bay is the major South African export port. Northwest Europe is the marker price for the Amsterdam-RoƩerdam-Antwerp region. Analysis is based on prices expressed in real terms. Sources: McCloskey Coal Report databases and IEA analysis.

154

World Energy Outlook 2013 | Global Energy Trends

RelaƟvely high coal prices in the period 2007-2011 were paralleled by cost escalaƟons driven by exchange rate eīects as well as rising costs for mining materials, consumables and labour as a result of the global mining and resources boom. Furthermore, healthy margins had taken pressure oī mining companies to increase producƟvity and cut costs in that period. Finally, strong coal demand kept high-cost mines in the market that were only marginally proĮtable, even at high prices. Consequently, falling prices since mid-2011 have squeezed margins and put pressure on the coal industry worldwide.

1

Currently, high costs and ample supply capacity is driving consolidaƟon and restructuring of mining industries in many countries. The United States experienced a wave of consolidaƟon starƟng in 2012 with large-scale producƟon cuts and mine closures, mainly in the Appalachian basins. However, since internaƟonal coal prices dropped below $80ͬtonne for all key exporters in the summer of 2013, other countries have been aīected as well. Australian and Canadian companies have reduced their workforce, idled unproĮtable mines and revised or delayed investment in new mines and infrastructure. At current prices even some Indonesian producers are considering producƟon cuts and a reducƟon of the workforce. Many coal companies have reacted to the plummeƟng prices with output expansion, trying to maintain revenues, causing prices to drop further. Given the current cost structure in the internaƟonal market, prevailing price levels are not sustainable in the long run without further reducƟons in high-cost supply.

4

Rampant growth in coal demand and imports means that Asia is increasingly the focus of coal markets. Qinhuangdao, the world’s largest coal port and China’s primary transhipment hub, has rapidly developed into a key pricing point in internaƟonal steam coal trade. While Europe is sƟll a large coal importer, accounƟng for one-ĮŌh of global steam coal imports, currently prices there tend to be lower, because of over-supply in the AtlanƟc Basin. Russia, the world’s third-largest steam coal exporter, is able to swing some of its exports between the AtlanƟc and PaciĮc markets and, in recent years, growing volumes have been sold to Asian buyers.9 In 2011, these volumes amounted to some 30% of Russian steam coal exports, with Japan and
© OECD/IEA, 2013

CompeƟƟon with natural gas in the power sector will be pivotal to coal price formaƟon. Anywhere that a CO2 price is in place, coal use and coal-Įred power will be aīected, to an extent that depends on the stringency of the CO2 price. In the European Union, the largest region with explicit carbon pricing under a cap-and-trade system, CO2 prices have been low for several years and fell further in 2012 and 2013. In the summer of 2013, they stood at less than $6ͬtonne. This, coupled with low coal prices in the AtlanƟc Basin, meant that coal was oŌen the lowest-cost fuel choice in the power sector. At coal and gas price levels prevailing in Europe in early 2013, the CO2 price would need to increase to close to $60ͬtonne to enable even a highly-eĸcient gas-Įred power plant to compete against a 1980s coal-Įred power staƟon.

2 3

5 6 7 8 9 10 11 12 13 14 15 16 17 18

9.ഩ See Medium-Term Coal Market Report 2013 for an in-depth analysis of the Russian coal market (IEA, 2013b).

Chapter 4 | Coal market outlook

155

Other factors will also aīect future coal price formaƟon. High cost exporters, such as Australia, the United States and Russia, may limit supplies to the internaƟonal market if cost escalaƟon conƟnues, puƫng upward pressure on internaƟonal coal prices. In addiƟon, seaborne bulk freight rates, which have been low since they fell sharply in 2008, could rise in the longer term. That would favour suppliers closer to the main Asian markets, such as Indonesia and Australia, at the expense of exporters from South Africa and the United States. A weaker US dollar would increase local costs (expressed in US dollar terms) in countries with a high proporƟon of locally sourced inputs, while a strengthening US dollar would tend to reduce them. SigniĮcant devaluaƟon of currencies in imporƟng countries, such as in India recently, will limit their ability to pay higher prices.

Regional insights China As the world’s largest coal user, producer and now importer, China dominates the global coal market. It maintains this pivotal role in the New Policies Scenario even though some trends are set to change markedly. China’s economic success has been fuelled primarily by coal, which provides over two-thirds of China’s primary energy demand. The country now uses nearly twice as much coal as all OECD countries combined.

© OECD/IEA, 2013

Coal is the backbone of China’s power system, comprising almost 80% of electricity generaƟon. zet, the power sector only accounts for about 55% of China’s coal consumpƟon, a much lower share than in other major coal users, like the United States, European Union and India. The remainder of China’s coal consumpƟon is in the iron and steel, and cement industries and in burgeoning applicaƟons, such as petrochemical feedstocks. Even the buildings and agriculture sectors use substanƟal amounts of coal. Nevertheless, developments in the power sector will be criƟcal to the prospects for Chinese coal demand and imports. In the period 2002-2011, electricity generaƟon in China nearly tripled, with about 80% of the incremental output coming from coal. In the period to 2020, electricity generaƟon in China increases by a further 55% in the New Policies Scenario, but only one-third is coal-Įred (Figureථ4.7). Renewables, especially hydro but also wind, supply more than 40% of incremental output, generaƟng an addiƟonal 1ථ075ථterawaƩ-hour (TWh) by 2020, compared with 2011 (roughly equal to the current electricity output of Japan). A signiĮcant contribuƟon also comes from nuclear, with 29 reactors currently under construcƟon in addiƟon to the 17ථunits already in operaƟon. As a result, growth in coal use in the power sector slows markedly, from 10.5% per year on average over 2002-2011 to 1.5% per year over 2011-2020. AŌer 2020, the growth in power sector demand falls to 0.8% per year, as the power mix is diversiĮed further and eĸciency gains curb the growth in electricity demand. However, the massive stock of coal-Įred staƟons already in place, combined with the country’s enormous coal reserves, ensures that coal remains the key source of electricity generaƟon, despite its share falling from 79% in 2011 to 55% in 2035.

156

World Energy Outlook 2013 | Global Energy Trends

TWh

Figure 4.7 ‫ ٲ‬China’s electricity generation in the New Policies Scenario

1

8 000

Gas

7 000

Nuclear

6 000

Other renewables

5 000

2

Hydro Growth 2011-2020

4 000

Coal

4

3 000

5

2 000 Growth 2002-2011

1 000

6

2020

2002

Coal is also the dominant fuel in China’s industry sector. Coal consumpƟon in industry (including coke ovens, blast furnaces, petrochemical feedstocks and coal-to-liquids) is greatest in the iron and steel industry, which accounts for over 50%, while the cement industry accounts for 20%. Over the period 2002-2011, industrial coal demand grew very quickly, averaging 9.5% per year. In the New Policies Scenario, growth is projected to slow signiĮcantly to 1.9% per year in the period to 2020, as a result of slower industrial growth and structural transformaƟon of the economy away from heavy industries (Figureථ4.8). The producƟon of crude steel and cement peaks before 2020 and declines thereaŌer, which results in a fall in industrial coal use of 0.9% per year in the period 2020-2035. Wider deployment of alternaƟve technologies in steel producƟon, such as electric arc furnaces, and achievements in eĸciency improvements reduce coal demand. Figure 4.8 ‫ ٲ‬Coal demand in China and the rest of the world by major sector in the New Policies Scenario Mtce

3

2035

2 000

Change 2011-2035 Change 1987-2011

1 200

8 9 10 11 12 13

2 400

1 600

7

1987

14 15

800

16

400 China

Rest of world

© OECD/IEA, 2013

Power generaon

China

Rest of world

17

Industry*

*ථIncludes own use and transformaƟon in blast furnaces and coke ovens, coal-to-liquids, and petrochemical feedstocks.

Chapter 4 | Coal market outlook

157

18

By contrast, both coal-to-liquids and coal use as feedstock in the chemical industry grow strongly, the laƩer by 4.7% annually in the period to 2035, driven by increasing demand for methanol. In the second half of the Outlook period, a substanƟal amount of domesƟcally produced methanol is directed towards the producƟon of oleĮns in the petrochemical industry, making use of the methanol-to-oleĮn process. Given the coal, oil and gas prices derived from IEA World Energy Model, producƟon economics favour coalbased petrochemical producƟon over that based on oil and gas, helping to reduce China’s dependence on imports. In previous years, rapidly increasing coal demand put strains on coal supply. To address this, China’s 11thථFive-zear Plan, issued in 2007, included mining industry reform that targeted producƟvity and safety improvements, as well as capacity expansion. In a wave of industry consolidaƟon, more than 9ථ000 small coal mines have been closed or incorporated into large mining complexes (IEA, 2012). The reforms have been successful, improving safety and producƟvity, and increasing coal mining capacity. Although coal mines are scaƩered across China, three provinces located in the north and northeast – Inner Mongolia, Shanxi, and Shaanxi – produce around 60% of the country’s coal. China’s coal producƟon grows by 0.7% per year between 2011 and 2020 in the New Policies Scenario, but stabilises thereaŌer, reŇecƟng slower growth in power sector coal use and declining coal use in heavy industry. The recent slowdown in growth has squeezed the proĮtability of coal mining in China, especially in old high-cost mines, increasing incenƟves to cut costs and triggering discussions about a possible ban on certain coal imports of low caloriĮc value, leading to the introducƟon of a modest import tax on such coals in August 2013. Since Indonesian coals come under China’s free-trade agreement with ASEAN countries, they are not aīected, so the immediate impact of the tax seems likely to be small. However, it may serve as a warning to potenƟal investors considering the development of projects for export to China and elsewhere. Other measures are also being considered to help domesƟc coal producers, including tax rebates.

© OECD/IEA, 2013

The key challenge in China’s coal market is the geographic mismatch of supply and demand, which requires coal to be hauled long distances. While more than 80% of Chinese steam coal output can be produced at less than $60ͬtonne, producƟon costs in some older operaƟons in Shanxi are closer to $80ͬtonne. Adding transport costs to southern or coastal demand centres can make some Chinese coal relaƟvely expensive (Figureථ4.9). This has opened the door to growing imports. China’s coal imports reached a record level of about 220ථMtce in 2012 and have conƟnued to grow strongly in the Įrst half of 2013. In the New Policies Scenario, imports are projected to peak within the current decade, then to decline slowly, though they remain, on average, above 2012 levels over the projecƟon period. Fierce compeƟƟon will persist between domesƟc supply and imports in southern and coastal China over the Outlook period. In China’s more mature mining areas, costs are projected to rise, encouraging new mines to open further west, where mining costs are low, but transport distances are longer.

158

World Energy Outlook 2013 | Global Energy Trends

Dollars per tonne (adjusted to 6 000 kcal/kg)

Figure 4.9 ‫ ٲ‬Average costs of steam coal delivered to coastal China, 2012 120

Import tax/VAT**

100

Seaborne freight to coastal China

80

Inland transport

2 3

Other costs*** 60

4

Mining costs

40

Average import price

5

20

Australia

Russia

China

United States*

1

6

Indonesia

*ථUnited States refers to exports from the Powder River Basin only. ** VAT is value-added tax. *** Includes coal preparaƟon, mine overhead, port charges, royalƟes and other taxes.

7

Sources: Wood Mackenzie databases and IEA analysis.

8

United States The United States is the world’s second-largest coal consumer, making up 13% of the global market, and by far the largest in the OECD, accounƟng for 45% of OECD coal use.10 In 2011, coal met around one-ĮŌh of US energy needs, based on large and, in many cases, easily mined and low-cost reserves. TradiƟonally, coal has fuelled more than half of electricity producƟon in the United States. However, over the past two decades or so, gas-Įred plants have been preferred for new capacity, causing coal’s share of total electricity generaƟon to drop, from 53% in 1990 to 43% in 2011. Gas-Įred generaƟon was favoured in 2012, parƟcularly in eastern regions where coal prices are relaƟvely high, due to low-cost gas (from conƟnued growth in shale gas producƟon) and reduced electricity needs (from a historically warm winter). At one point in early spring 2012, gas and coal had almost the same share of electricity output, at around 33% each (US EIA,ථ2013). Since that Ɵme, gas prices have risen to around $3.8ͬMBtu in June 2013 and coal has regained some market share: it was back to almost 40% of total electricity, with gas falling to 28%.

© OECD/IEA, 2013

The power sector, as in most other countries, holds the key to US coal demand. In the New Policies Scenario, coal-Įred generaƟon declines both in relaƟve and absolute terms over the projecƟon period, despite a temporary rebound in the medium term. Power sector coal consumpƟon stood at nearly 625ථMtce in 2011 and is projected to drop to just under 520ථMtce in 2035. Furthermore, the US Ňeet of coal-Įred power plants is ageing and will be aīected by increasingly stringent environmental regulaƟons. One new set of regulaƟons, the Mercury and Air Toxics Standards, which are expected to be implemented by 2016, could lead to the closure of more than 20ථGW of coal10.ഩ See Medium-Term Coal Market Report 2012 for an in-depth analysis of the US coal market (IEA, 2012).

Chapter 4 | Coal market outlook

159

9 10 11 12 13 14 15 16 17 18

Įred capacity by some esƟmates (Beasley, et al., 2013).11 Recent announcements by the USථadministraƟon, instrucƟng the US Environmental ProtecƟon Agency (EPA) to draŌ carbon polluƟon standards for new and exisƟng coal-Įred power staƟons, if adopted, may have an even bigger impact (Boxථ4.4). The New Policies Scenario cauƟously implements emissions reducƟon strategies, and accordingly, coal-Įred capacity falls by 20%, from 335ථGW in 2011 to 265ථGW in 2035. US coal producƟon at 765ථMtce in 2011, aŌer a modest recovery in the medium term, is projected to enter a slow decline, echoing falling demand in the New Policies Scenario and reaching 655ථMtce in 2035. High-cost producers in the central Appalachia region (mainly southern West Virginia and eastern
Dollars per tonne (adjusted to 6 000 kcal/kg)

Figure 4.10 ‫ ٲ‬US production cash costs for domestic steam coal, 2012 120

Powder River Basin

100

Illinois Basin Northern Appalachia

80

Central Appalachia Other

60 40 20

0

100

200

300

400

500

600 700 Million tonnes

© OECD/IEA, 2013

Sources: Wood Mackenzie databases and IEA analysis.

11.ഩ The Cross-State Air Pollution Rule, a separate rule to the Mercury and Air Toxics Standards, that could have had a significant impact on coal-fired power plants, was ruled void in late 2012. 160

World Energy Outlook 2013 | Global Energy Trends

Box 4.4 ‫ ٲ‬US President’s Climate Action Plan

1

In June 2013, the US administraƟon outlined a climate acƟon plan, encompassing measures to reduce the country’s greenhouse-gas emissions, prepare the naƟon for the impacts of climate change and lead internaƟonal eīorts to combat climate change. In the absence of congressional agreement on climate policy, the plan relies mainly on execuƟve powers. One important measure covers carbon emissions from exisƟng and new power plants, potenƟally covering, inter alia, some 335ථGW of coal-Įred generaƟng capacity. The plan includes a mandate to the EPA to develop regulaƟons to control CO2 emissions from power plants, similar to exisƟng federal limits on emissions of arsenic, lead and mercury. The Įrst ever naƟonal standards for new power plants, which require them to limit CO2 emissions to 1ථ000ථpounds (454ථkg) per megawaƩ-hour (MWh), were proposed in March 2012. Following extensive public consultaƟon, a revised standard was issued in late September 2013, limiƟng new coal-Įred plants to emissions of 1ථ100ථpounds (499ථkg)ͬMWh. New gas-Įred plants were limited to emissions of 1ථ000ථpoundsͬMWh. Since even best pracƟce coal-Įred plants cannot produce electricity with emissions below 700ථkgͬMWh, the latest EPA provision eīecƟvely means that no new coalĮred plants can be built without a signiĮcant porƟon of emissions being captured and sequestered. Best pracƟce combined-cycle gas units can meet the EPA gas-Įred power emissions standard without CCS.

© OECD/IEA, 2013

With the standard set for new power plants, the EPA can be expected to issue guidance on standards for exisƟng power plants, based on wide ranging consultaƟons with a variety of stakeholders, and using a co-operaƟve approach with states. Each state will be asked to develop and submit plans to reduce emissions from exisƟng coal-Įred plants, using strategies that are Ňexible, account for regional diversity, and allow every available fuel source to conƟnue to be uƟlised. The Ɵming for this part of the iniƟaƟve is ambiƟous, with a Įrst proposal from EPA by June 2014, a Įnal one by June 2015 and state plans in 2016. Compliance acƟons could be required as soon as 2018. An alternaƟve compliance method, which would allow the averaging of emissions over a number of years, is also being considered, and depending on its Įnal form, could signiĮcantly alter the impact of the regulaƟons. The US President’s Climate AcƟon Plan is also designed to sƟmulate investment in advanced fossil energy projects, with up to $8ථbillion in loan guarantees to be oīered for innovaƟve technologies that can avoid, reduce or sequester CO2 emissions. The plan also includes working with other countries heavily dependent on coal-Įred power to speed up the development and deployment of clean coal technologies, as well as measures to promote coal-to-gas switching in the power sector.

2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18

Chapter 4 | Coal market outlook

161

The prospects for US coal producƟon depend on export demand as well as domesƟc needs. Net exports of coal in the New Policies Scenario are projected to remain around 80 Mtce for much of the projecƟon period, helping to compensate for falling domesƟc demand, before dipping to 65ථMtce by 2035. The share of net exports in total producƟon remains broadly Ňat at around 11%͖ however net exports of steam coal are quite modest aŌer 2015 (Figureථ4.11). Net exports of steam coal alone increased by over 80% to reach record levels of 40ථMtce in 2012 as steam coal, especially from the eastern states, was increasingly displaced by gas in the power sector. Much of the displaced coal went to Europe, where higher gas prices prevailed, making increased coal imports compeƟƟve in the power sector (see Chapterථ5, Boxථ5.1). The recent rapid growth of US exports, in combinaƟon with dwindling US coal imports from Colombia, has been a major contributor to ample supply on the internaƟonal steam coal market.

Mtce

Figure 4.11 ‫ ٲ‬US net exports of coal in the New Policies Scenario 100

20%

80

16%

60

12%

40

8%

20

4%

0

0%

Steam coal Coking coal Share of net exports in US producon (right axis)

-20 1990 1995 2000 2005 2010 2011 2012* 2020 2035

© OECD/IEA, 2013

* Preliminary data.

The regional origin of exported coal from the United States is expected to shiŌ gradually from the east to west in the New Policies Scenario. High costs and reserves depleƟon reduce steam coal exports from the Appalachian basins, while steam coal exports from the Illinois basin and the western United States grow moderately. For coking coal the situaƟon is diīerent, as exported coal comes exclusively from the Appalachian basins. Coal exports from the Powder River Basin are proĮtable under current market circumstances, but any large-scale expansion of these exports would require substanƟal capital investment in US or Canadian port faciliƟes on the west coast, as well as the upgrading of the exisƟng railway infrastructure. Public opposiƟon to coal exports through west coast ports is growing and is likely to delay projects. Moreover, due to the relaƟvely low energy content of the Powder River Basin coals and the long transport distance to Asian demand centres, the economics of coal exports are parƟcularly exposed to changes in freight rates and railway tariīs. Uncertainty about future export demand is an element of investment risk, while China’s current and potenƟal future policies towards restricƟng low caloriĮc-value coal imports 162

World Energy Outlook 2013 | Global Energy Trends

further exacerbate this risk. Nonetheless, the current price diīerenƟals between western US and internaƟonal coal prices provide a strong incenƟve to seek innovaƟve ways of exporƟng this coal to Asia (Figureථ4.9).

India The outlook for coal demand in India – the world’s third-largest coal consumer – contrasts starkly with that of China. As in China, coal has played a pivotal role in India’s recent economic growth – coal use doubled between 2000 and 2011 to 465ථMtce (and nearly 485ථMtce in 2012 based on preliminary data), with growth acceleraƟng in the laƩer half of this period. Two-thirds of demand now comes from the power sector, where coal accounts for 68% of electricity output. Industry makes up almost all the balance of coal demand, dominated by iron and steel, and cement. Unlike China, demand conƟnues to grow strongly in the New Policies Scenario, more than doubling by 2035. India overtakes the United States as the world’s second-largest coal user soon aŌer 2020͖ yet per-capita electricity use remains very low at that Ɵme, suggesƟng considerable potenƟal for further demand growth for electricity (and coal). It is esƟmated that in 2011, one out of four people in India (about 306ථmillion) did not have access to electricity, while two out of three (about 818ථmillion) relied on the tradiƟonal use of biomass for cooking (see Chapterථ2). Despite the doubling in coal use to generate power, coal’s share of electricity output declines, from 68% in 2011 to 56% in 2035, as renewables, nuclear and gas gain share. Around half of the new coal-Įred capacity installed over the projecƟon period is based on subcriƟcal technologies. Industrial coal demand also more than doubles to 2035, increasing its share in the sector, on the back of a tripling of crude steel output and a doubling of cement producƟon.

© OECD/IEA, 2013

India has a large, although generally poor quality, coal resource base of around 210ථbillion tonnes (BGR, 2012). Coal output expanded rapidly – by more than two-thirds – between 2000 and 2009. Output has barely increased since then, as the state-owned mining company, Coal India Limited, which dominates the sector, has faced a number of diĸculƟes, including lack of access to coal deposits, a lack of or mismatch of rail capacity and limited access to advanced mining technology. Many coal deposits are in populated or forested areas, necessitaƟng signiĮcant disturbance to ecosystems or the movement of large numbers of people if open-cut mining methods are to be used. Increasing the level of compeƟƟon in the Indian coal sector, as well as allowing for foreign investment, would introduce advanced mining technology, and facilitate an expansion of supply and higher producƟvity. The diĸculƟes in bringing new mining capacity online are expected to persist in the current decade, resulƟng in the majority of India’s projected 75% increase in coal output in the New Policies Scenario occurring aŌer 2020. With domesƟc coal output struggling to keep up with booming demand (which has resulted in coal-Įred capacity running well below technical limits in many regions), Indian imports have nearly doubled since 2008. India’s coal-import dependency has increased signiĮcantly over the past decade, a trend that is expected to persist over the projecƟon period. Imports Chapter 4 | Coal market outlook

163

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18

as a share of supply jumped from 9% in 2000 to 23% in 2011. According to preliminary data, imports grew a further 17% in 2012, resulƟng in India overtaking
Mtce

Figure 4.12 ‫ ٲ‬Major net importers of coal in the New Policies Scenario 400

2035

300

Change 2020-2035 Change 2011-2020

200

2011 100 0 -100

India

Rest of Asia

China

Japan

European Union

Korea

© OECD/IEA, 2013

The price of imported coal to Indian power generators will determine the level of demand. Private generators have recently been squeezed by long-term fixed prices under power purchase agreements coupled with rising costs from more expensive coal imports, compared with local output. A price pooling system has recently been agreed, whereby local and imported coal prices are rolled together and the resultant price applied uniformly to all generators. In addition, the government has also recently agreed to allow the pass- through of imported coal prices for power producers. While only the first steps in opening this market, these developments are likely to facilitate increased coal supply. Australia is the major supplier of coking coal to India, while Indonesia and South Africa are the main sources of steam coal. Indonesia is set to remain the largest steam coal supplier to India into the medium term, but Australia figures more prominently later in the projection period. To satisfy the growing need for imports, Indian investors are acquiring resources overseas, including in Australia, Mozambique and Indonesia. For example, in Australia, large steam coal projects supported by Indian investors are planned for the Galilee Basin in Queensland, with an annual export capacity of some 120ථMt (BREE, 2013).

164

World Energy Outlook 2013 | Global Energy Trends

Australia Australia is now the second-largest coal exporter in the world, having been overtaken by Indonesia in 2012. It is set to conƟnue to vie for the top spot over the projecƟon period, regaining it around 2030 in the New Policies Scenario. Based on preliminary data, Australian coal exports jumped almost 6% in 2012, to nearly 280ථMtce, about half of which was coking coal (Australia sƟll leads coking coal exports, supplying over 50% of all internaƟonally traded coking coal). Australian coal exports are projected to conƟnue to increase substanƟally, to 410ථMtce in 2035 – up 57% from 2011 levels (Figureථ4.13). Steam coal exports grow by three-fourths to around 225ථMtce, with most of the increase expected to come from the as yet undeveloped Galilee Basin in central Queensland, while coking coal exports grow by nearly 40%, to 190ථMtce. Australia’s share of internaƟonal trade in steam and coking coal rises by Įve percentage points (to 24%) and one percentage point (to 54%), respecƟvely, largely at the expense of the United States. The main desƟnaƟons for Australian steam coal are Japan,
Mtce

90%

Coking coal

400

80%

Steam coal

350

70%

Share of world trade (right axis):

300

60%

250

50%

200

40%

150

30%

100

20%

50

10% 1990

2000

2011

2020

© OECD/IEA, 2013

3 4 5 6 7

9 10

Coking coal Steam coal

11 12 13

2035

A few years ago, high producƟvity, favourable geological condiƟons and good coal quality meant that Australian coal exporters were posiƟoned roughly half way along the global supply cost curve for internaƟonally traded steam coal – despite comparaƟvely high labour costs. Today, the costs of Australian mines are among the highest (Figureථ4.5). This change has come about as a result of deterioraƟng mining condiƟons, escalaƟng labour costs, the increased costs of mining materials and appreciaƟon of the Australian dollar.12 The Įrst two of these eīects are a direct or indirect result of the resources and minerals boom that has been a feature of the Australian economy in recent years. Cuƫng costs to remain 12.ഩ The importance of the exchange rate is illustrated by a simple calculation: coal sales at $100ͬtonne yield A$128ͬtonne at the 2009 exchange rate of $1 с A$1.28, but at the 2012 rate of A$0.97, only A$97ͬtonne, slashing revenues needed to pay costs in local currency (such as labour or transport) by more than A$30.

Chapter 4 | Coal market outlook

2

8

Figure 4.13 ‫ ٲ‬Australian coal exports by type in the New Policies Scenario 450

1

165

14 15 16 17 18

compeƟƟve in the internaƟonal market is a key challenge for Australian coal exporters over the Outlook period. A recent fall in the exchange rate by around 15%, indicaƟng a decline in the value of the Australian dollar, will certainly improve proĮtability, but the currency remains well above its historical level, with conƟnuing impacts on investment prospects. Investment in Australia’s resource sector has seen a massive boom over recent years. While liqueĮed natural gas (LNG) projects have played a large part (see Chapterථ3), expanding producƟon of steam and coking coal has also been signiĮcant. However, new coal mines have become more capital-intensive in recent years, due to escalaƟng costs for mining equipment and construcƟon labour. Australia is now one of the most expensive places to build a new mine. Even so, 93 coal projects are planned, with total capacity of up to 590ථMt per year, although only 16 of these, with a capacity of some 60ථMt per year, have been commiƩed (BREE, 2013). Of greatest interest among the uncommiƩed projects are Įve large, greenĮeld steam coal mines in the Galilee Basin (with capaciƟes totalling nearly 180ථMt), three of which have signiĮcant Indian involvement.

ASEAN13 Coal use in ASEAN countries is poised to triple, driven by rapid economic and populaƟon growth (see Chapterථ2). The power sector will be the principal driver of coal demand. In the New Policies Scenario, electricity demand more than doubles from around 700ථTWh in 2011 to 1ථ880 TWh in 2035, growth equivalent to more than Japan’s electricity output in 2010. Mirroring the trend seen in China and India, coal is emerging as the fuel of choice for power generaƟon in the ASEAN region. Coal already accounts for some three-quarters of thermal power generaƟon capacity under construcƟon in ASEAN countries, resulƟng in coal’s share of the electricity mix growing rapidly from around 30% to 49% over 20112035, mainly at the expense of natural gas and oil. This boosts ASEAN coal demand from 130ථMtce in 2011 to 400ථMtce in 2035, a growth rate of 4.8% per year – the fastest of any major coal-consuming region or country (Figureථ4.14). Coal use in ASEAN exceeds that in the European Union before 2030.

© OECD/IEA, 2013

ASEAN coal demand growth is supported by the relaƟve abundance and low cost of local resources and, as yet, a lack of stringent environmental standards. At the same Ɵme, gas faces increasing development costs, someƟmes low domesƟc prices and the possibility of realising higher value through LNG exports. One important trend is that the majority of coal-Įred power plants under construcƟon or planned employ subcriƟcal technology, locking in low eĸciency technology for decades to come. Such plants use up to 15% more coal for a given power output than more eĸcient supercriƟcal plants, which are Įnancially viable at current coal prices and capital costs (IEA, 2013c). Indonesia, the country with the world’s fourth-largest populaƟon and ASEAN’s largest energy user, is set to lead the growth in the region’s coal demand, with its abundant coal resources and an already booming export sector. Coal demand in Indonesia has increased 13.ഩ ASEAN energy prospects are analysed in-depth in the Southeast Asia Energy Outlook, a WEO special report (IEA, 2013c). 166

World Energy Outlook 2013 | Global Energy Trends

at 11% per year for the last two decades. It more than triples in the New Policies Scenario, reaching 165ථMtce in 2035, exceeding the current level of use in Japan. The government plans to meet rapidly rising power demand through a large expansion of coal-Įred power generaƟon. It has given priority to increasing coal supply to the domesƟc market vis-àvis exports. Consequently, the government has set a minimum share of coal producƟon that must be sold to domesƟc customers. Also, the government has discussed banning low quality coal exports (less than 5ථ600ථkilocalories per kilogramme) to ensure fuel supply for new coal-Įred power plants (though formal regulaƟons have not been adopted).

Mtce

Figure 4.14 ‫ ٲ‬ASEAN coal balance in the New Policies Scenario

2 3 4 5

800

Net exports: Indonesia

600

Net imports: Rest of ASEAN

6 7

Producon

400

1

Demand

8

200 0

9

-200 2000

2005

2011

2020

2025

2030

10

2035

Thailand, the region’s second-largest energy user, currently has a power sector dominated by gas, but interest in coal-Įred power is increasing, mainly in response to energy security and cost issues. In Malaysia, coal also grows in importance as the rise in gas demand outstrips indigenous gas supply growth. In the Philippines, coal accounts for two-thirds of incremental power generaƟon over 2011-2035 (IEA, 2013c).

© OECD/IEA, 2013

In the New Policies Scenario, the ASEAN region sees a conƟnuaƟon of strong growth in coal producƟon in the medium term with output rising to 510ථMtce in 2020. Indonesia is the main source of coal to meet both strong growth in domesƟc use and demand for steam coal exports to the Asia-PaciĮc market. Over the long term, producƟon from the region slows as coal producers face rising costs in developing their resources, as a result of higher costs for mining, labour and transport, which undermine the compeƟƟveness of exports. Excluding Indonesia, ASEAN countries as a group see coal imports rise more than Įve-fold over the projecƟon period, exceeding Japan’s current import levels. A signiĮcant increase in Vietnam’s coal-Įred power generaƟon in the short term, to improve electricity reliability, is set to shiŌ the country from being a net exporter to a net importer of coal, though the extent and Ɵming depends on the eĸciency of generaƟng technologies chosen for these plants, as well as the pace at which they are commissioned. FaciliƟes are being developed that would receive coal imports from Australia, Indonesia and Russia. Chapter 4 | Coal market outlook

167

11 12 13 14 15 16 17 18

© OECD/IEA, 2013

Chapter 5 Power sector outlook Capacity to change? Highlights

x Demand for electricity grows more than demand for any other Įnal form of energy. In the New Policies Scenario, world electricity demand increases by more than twothirds over the period 2011-2035, growing at an average rate of 2.2% per year. It is driven by increasing electriĮcaƟon of industry, where electricity’s share grows from 26% to 32%, expanded use of electrical appliances and more cooling in buildings.

x Non-OECD countries account for the greater part of incremental electricity demand by far, led by China (36%), India (13%), Southeast Asia (8%), LaƟn America (6%) and the Middle East (6%). In terms of electricity demand per capita, the gap narrows between non-OECD and OECD countries, but only Russia, China and the Middle East exceed even half of the OECD average in 2035͖ sub-Saharan Africa reaches just 6% of the OECD average at the end of the projecƟon period.

x Global installed generaƟng capacity grows by over 70%, from 5ථ649ථGW in 2012 to about 9ථ760ථGW in 2035, aŌer reƟring some 1ථ940ථGW of generaƟng capacity. About 60% of reƟrements are in OECD countries, where about two-thirds of the coal Ňeet is already over 30 years old. China’s addiƟons of coal, nuclear (more than current nuclear capacity in the United States) and renewables are the most of any region. AddiƟons of renewables in the European Union are the second-largest globally.

x Though it remains the leading fuel, coal’s share of generaƟon falls from 41% to 33%. The share of renewables rises from 20% to 31% while the shares of gas and nuclear hold steady at 22% and 12% respecƟvely. Though total CO2 emissions increase, greater use of lower-carbon sources and more eĸcient fossil-fuelled plants contribute to a 30% drop in the CO2 emissions intensity of the power sector.

x The length of transmission and distribuƟon (TΘD) lines expands from 69ථmillion km in 2012 to 94ථmillion km in 2035, largely in response to fast-growing electricity demand in non-OECD countries, including that of new end-users. By 2035, around 50% of today’s grid infrastructure will have reached 40ථyears of age, necessitaƟng major investments in refurbishment during the projecƟon period.

x SubstanƟal investments in the power sector will be required over the projecƟon

© OECD/IEA, 2013

period to saƟsfy rising electricity demand and to replace or refurbish ageing infrastructure. CumulaƟve global investment in the power sector is $17.0ථtrillion over 2013-2035, averaging $740ථbillion per year. New plants account for 58% of the total, while the rest is needed in TΘD networks.

x Electricity prices rise in most regions, with widening regional diīerences. By 2035, US industrial electricity prices are half the level in the European Union and 40% lower than those in China, which could have important ramiĮcaƟons for compeƟƟveness. Chapter 5 | Power sector outlook

169

IntroducƟon The power sector is a complex amalgam of thousands of power plants, millions of kilometres of lines in the transmission and distribuƟon network and billions of end-users, with system operators balancing demand and supply in real Ɵme. Many factors inŇuence the pace of electricity demand growth such as gross domesƟc product (GDP), electricity prices, populaƟon growth, the proporƟon of populaƟon with access to electricity supply, standards of living, and the extent of the deployment of energy-eĸcient equipment. The evoluƟon of the mix of generaƟng plants to meet demand depends largely on the relaƟve economics of diīerent energy technologies, taking account of fossil-fuel prices, carbondioxide (CO2) pricing (if applicable), the capital costs of power plants, Įnancing condiƟons, policies to promote or limit the deployment of speciĮc technologies, the availability of domesƟc fuel resources, the age of the exisƟng power plant Ňeet and the structure of the power market. Which power plants are run to meet electricity demand typically depends on the variable costs of their operaƟon. Plants with the lowest variable costs are generally dispatched Įrst, however, much depends on how the local power market is organised. There are two basic designs: fully liberalised markets and fully regulated systems though, in pracƟce, most systems have some features of both designs. Worldwide, most power is generated in relaƟvely highly regulated systems. The design of the system determines how prices are formed and the condiƟons for investment. Policy intervenƟons have to be tailored to the design of the individual system. The last twelve months have seen signiĮcant developments in a number of electricity markets around the world. For example, in the United States, excepƟonally low gas prices in 2012 led to a strong surge in gas-Įred electricity generaƟon, displacing coalĮred generaƟon. The opposite was true in the European Union: as natural gas became increasingly expensive, compared to coal, this – in combinaƟon with low CO2 prices, weaker economic acƟvity, lower electricity demand and conƟnued expansion of renewablebased capacity – led to a noƟceable drop in gas-Įred generaƟon in 2012 compared to the previous year. Europe has also seen conƟnued strong growth of variable renewables that have increasingly impacted the operaƟon of convenƟonal power plants and lowered wholesale power prices in some systems.

© OECD/IEA, 2013

Japan saw a surge in renewables capacity, in response to new support policies put in place, in parƟcular for solar photovoltaics (PV), designed to curtail the sharp rise in the cost of imporƟng fuels for oil- and gas-Įred generaƟon following the substanƟal reducƟon of nuclear power generaƟon aŌer the Fukushima Daiichi accident. In
170

World Energy Outlook 2013 | Global Energy Trends

for nuclear plants was liŌed and construcƟon of new plants restarted, although at a slower pace than before the Fukushima Daiichi accident. An ambiƟous new target for 2015 for solar PV was introduced towards the end of the year. In India in 2012, low availability of gas and lower than planned capacity addiƟons (partly due to unfavourable market frameworks) Ɵghtened the demand-supply balance. Similar Ɵghtness was experienced in Brazil, due to low hydropower reservoir levels, resulƟng in a strong call on fossilථfuel-Įred power plants throughout the year. Southeast Asia saw a conƟnued push towards investment in new coalĮred power plants, aimed either at reducing costly gas-Įred generaƟon, or boosƟng gas exports (in gas-producing countries).

1

Electricity demand

5

World electricity demand nearly doubled between 1990 and 2011, growing at an average rate of 3.1% per year. Between 2011 and 2035, demand for electricity grows more than any other Įnal form of energy in all of the three scenarios analysed in this Outlook. Electricity demand is strongly linked to future economic growth, the overall level of which tends to be reŇected in the level of economic acƟvity in key electricity-consuming sectors (such as in industry and services). The rate of electricity demand growth in the three scenarios depends primarily on the nature and extent of government intervenƟons, parƟcularly policies related to energy eĸciency, the environment and energy security. Some of these policies inŇuence electricity demand directly, such as measures to improve end-use eĸciency and to encourage fuel switching, and some indirectly, through their impact on Įnal prices. In the New Policies Scenario, the central scenario in this Outlook, world electricity demand increases by more than two-thirds over the period 2011-2035, growing at an average rate of 2.2% per year. Demand rises more quickly in the Current Policies Scenario (2.5% per year) and more slowly in the 450 Scenario (1.7% per year) (Tableථ5.1).

© OECD/IEA, 2013

In the three scenarios, diīering policy measures and electricity prices determine the rate of uptake of more energy-eĸcient technologies and overall rates of improvement in energy eĸciency. In 2035, world demand is projected at 32ථ150ථterawaƩ-hours (TWh) in the New Policies Scenario, with variaƟons 12% below this Įgure and 7% above it. Taking electricity intensity – electricity consumpƟon per unit GDP – as a broad indicator of energy eĸciency1 in electricity end-uses, the New Policies Scenario sees an average annual rate of improvement of 1.1% during the projecƟon period, whereas energy eĸciency advances more slowly in the Current Policies Scenario (0.8% per year) and more quickly in the 450ථScenario (1.6% per year). Without any change in electricity intensity with respect to the last Įve years, world electricity demand would rise to 43ථ100 TWh in 2035. In the New Policies Scenario, it is lower by about one-quarter (Figureථ5.1). Energy eĸciency is the main driver of this diīerence (Chapter 7), although the shiŌ towards less energy-intensive sectors plays an important role too. 1.ഩ For any country, measuring energy efficiency is challenging as it requires extensive data collection and analysis. As an imperfect proxy, the electricity intensity gives a broad indication of strides towards improvements in energy efficiency, but it is important to note that each country will have significantly different electricity intensities based on factors such as level of industrialisation and climate.

Chapter 5 | Power sector outlook

171

2 3 4

6 7 8 9 10 11 12 13 14 15 16 17 18

Table 5.1 ‫ ٲ‬Electricity demand* by region and scenario (TWh) New Policies

Current Policies

450 Scenario

1990

2011

2035

20112035**

OECD

6 591

9 552

11745

0.9%

12 369

1.1%

10 934

0.6%

Americas

3 255

4 694

5 912

1.0%

6 103

1.1%

5 457

0.6%

United States

2035

20112035**

2035

20112035**

2 713

3 883

4 753

0.8%

4 883

1.0%

4 438

0.6%

Europe

2 320

3 160

3 740

0.7%

4 040

1.0%

3 564

0.5%

Asia Oceania

1 016

1 698

2 093

0.9%

2 226

1.1%

1 912

0.5%

758

954

1 119

0.7%

1 195

0.9%

993

0.2%

Non-OECD

Japan

3 493

9 453

20 405

3.3%

22 084

3.6%

17 323

2.6%

E. EuropeͬEurasia

1 584

1 367

2 004

1.6%

2 171

1.9%

1 730

1.0%

Russia

909

838

1 256

1.7%

1 375

2.1%

1 075

1.0%

1 049

5 888

13 913

3.6%

15 211

4.0%

11 758

2.9%

China

558

4 094

8 855

3.3%

10 023

3.8%

7 417

2.5%

India

Asia

212

774

2 523

5.0%

2 582

5.2%

2 198

4.4%

Middle East

190

702

1 484

3.2%

1 587

3.5%

1 216

2.3%

Africa

262

584

1 296

3.4%

1 304

3.4%

1 094

2.7%

LaƟn America

407

912

1 708

2.6%

1 811

2.9%

1 525

2.2%

214

471

939

2.9%

1 001

3.2%

834

2.4%

10 085

19 004

32 150

2.2%

34 454

2.5%

28 256

1.7%

2 241

2 852

3 246

0.5%

3 512

0.9%

3 120

0.4%

Brazil World European Union

* Electricity demand is calculated as the total gross electricity generated less own use in the producƟon of electricity, less transmission and distribuƟon losses. ** Compound average annual growth rate.

Figure 5.1 ‫ ٲ‬World electricity demand by scenario relative to electricity TWh

demand assuming no change in electricity intensity 45 000

Based on electricity intensity 2007-2011

40 000

-20% -25%

35 000 30 000

-34%

Current Policies Scenario New Policies Scenario

25 000

450 Scenario

20 000 15 000

© OECD/IEA, 2013

10 000 5 000 2020

172

2035

World Energy Outlook 2013 | Global Energy Trends

In the New Policies Scenario, industry maintains its posiƟon as the largest consumer of electricity throughout the Outlook period, accounƟng for 41% of total electricity demand in 2035. Industry demand growth averages 2.2% per year during the projecƟon period (Tableථ5.2), underpinned by increasing electriĮcaƟon of industrial processes, with electricity increasing its share of total energy supply to the sector from 26% to 32%. Demand in the residenƟal sector expands at 2.5% per year, more than two-and-a-half Ɵmes faster than the rate of populaƟon growth, reŇecƟng increased use of electrical appliances, more cooling and improved access to electricity. The share of the world populaƟon without access to basic electricity services falls from 18% (1.2ථbillion) in 2011 to 12% (970ථmillion) in 2030 (see Chapterථ2). Demand in the services sector increases more slowly, by 1.9% per year, the slower rate of growth reŇecƟng energy eĸciency measures in OECD countries and the direct use of renewables for heat. Electricity demand in the transport sector is the fastest-growing (averaging 3.9% per year), due to a doubling of electricity demand from rail. However, transport accounts for only just over 2% of total electricity demand in 2035, despite the inroads made by electric vehicles between 2011 and 2035, and the associated demand increasing by about 30% per year (though from a low level).

1 2 3 4 5 6 7 8

Table 5.2 ‫ ٲ‬World electricity demand by sector and generation in the New Policies Scenario (TWh) 1990

2011

2020

2025

2030

2035

20112035**

Demand

10 085

19 004

24 249

26 974

29 520

32ථ150

2.2%

Industry

4 419

7 802

10 288

11 385

12ථ268

13ථ187

2.2%

ResidenƟal

2 583

5 195

6 507

7 362

8ථ325

9ථ336

2.5%

Services

2 086

4 560

5 636

6 214

6ථ698

7ථ137

1.9%

Transport

245

292

408

486

590

734

3.9%

Other sectors

748

1 151

1 419

1 535

1ථ648

1ථ763

1.8%

1 003

1 816

2 308

2 589

2ථ862

3ථ138

2.3%

TΘD losses PG own use Gross generaƟon*

733

1 298

1 434

1 550

1ථ668

1ථ791

1.4%

11 818

22 113

27 999

31 121

34ථ058

37ථ087

2.2%

© OECD/IEA, 2013

*ථGross generaƟon includes demand in Įnal uses (industry, residenƟal, services, transport and other), losses through transmission and distribuƟon (TΘD) grids, and own use by power generators (PG). **ථCompound average annual growth rate.

Non-OECD countries account for by far the greater part of incremental electricity demand, driven by faster economic and populaƟon growth, shiŌs from rural to urban living and rising standards of living. In the New Policies Scenario, the largest sources of addiƟonal global demand are China (36%), India (13%), Southeast Asia (8%), LaƟn America (6%) and the Middle East (6%). Electricity demand growth in China abates considerably: having averaged 12% per year over 2000-2011, it slows to 3.3% per year over 2011-2035 (with slowing economic growth and a restructuring of the economy towards less energy-intensive sectors). Demand increases most rapidly in India (5.0% per year) and Southeast Asia (4.2%). In terms of electricity demand per capita, the gap narrows between non-OECD and OECD Chapter 5 | Power sector outlook

173

9 10 11 12 13 14 15 16 17 18

countries, but among developing countries, only China and the Middle East exceed even half the OECD average in 2035 (Figureථ5.2). The level remains very low in sub-Saharan Africa, at 520ථkWh, or just 6% of the OECD average, in 2035. Figure 5.2 ‫ ٲ‬Electricity demand per capita in selected regions as a share of the OECD average in the New Policies Scenario 2011

China

2035

Middle East Brazil Southeast Asia India Africa 10%

20%

30%

40%

50%

60%

70%

80%

Note: In the New Policies Scenario, average electricity demand per capita in OECD countries grows from 7ථ670 kWh in 2011 to 8ථ500 kWh in 2035.

Electricity supply World electricity generaƟon increases in line with incremental growth in electricity demand in each of the scenarios.2 The mix becomes more diverse, though the nature of its evoluƟon varies by region according to government policies and compeƟƟon between generaƟon types. The scenarios diīer most with respect to the pace of the transiƟon from fossil-fuelled to low-carbon generaƟon (Tableථ5.3). This depends criƟcally on the Ɵming and the rigour of policies adopted to address environmental concerns (local polluƟon and CO2 emissions) – which arise earliest and are strongest in the 450ථScenario (see Chapterථ2) and are more limited in the Current Policies Scenario than in the New Policies Scenario – as well as on relaƟve investment costs and fuel prices for diīerent generaƟng technologies.

© OECD/IEA, 2013

In the New Policies Scenario, world electricity generaƟon increases from 22ථ113ථTWh in 2011 to almost 37ථ100ථTWh in 2035 (or by two-thirds), growing at an average rate of 2.2% per year. Fossil fuels conƟnue to have a dominant role, although their combined share declines from 68% to 57%: coal remains the largest source of electricity generaƟon, growing steadily at around 1.2% per year. Natural gas expands most by almost 3ථ500ථTWh. Amongst the renewable energy technologies, increases in generaƟon from hydropower and wind, about 2ථ300ථTWh each, are highest, with renewables as a group accounƟng for almost half of the increase in global electricity generaƟon over 2011-2035. 2.ഩ In each of the scenarios the rate of growth for electricity generation is actually slightly lower than that for demand, reflecting falling shares of transmission and distribution losses, and own use by power generators. 174

World Energy Outlook 2013 | Global Energy Trends

Table 5.3 ‫ ٲ‬Electricity generation by source and scenario (TWh) New Policies ථ

Current Policies

1

450 Scenario

1990

2011

2020

2035

2020

2035

2020

2035

OECD

7 629

10 796

11 827

13 104

11 990

13 835

11 415

12 123

Coal

3 093

3 618

3 529

2 775

3 681

3 835

2 961

1 116

Gas

770

2 630

2 855

3 398

2 979

3 710

2 813

2 307

Oil

697

345

149

84

153

92

126

44

Nuclear

1 729

2 087

2 300

2 412

2 273

2 246

2 355

2 826

Hydro

1 182

1 388

1 490

1 615

1 476

1 586

1 523

1 730

Other renewables

157

728

1 504

2 820

1 428

2 367

1 637

4 099

Non-OECD

4 189

11 317

16ථ172

23 983

16 799

26 018

15 139

20 173

Coal

1 333

5 522

7 089

9 537

7 901

12 296

6 043

3 544

Gas

960

2 217

3 128

4 915

3 242

5 463

2 958

3 686

Oil

635

717

652

472

666

522

578

278

Nuclear

283

497

1 100

1 881

1 049

1 668

1 191

3 011

Hydro

963

2 102

3 065

4 212

2 936

3 891

3 144

4 665

Other renewables

15

263

1 138

2 965

1 004

2 177

1 225

4 989

11 818

22 113

27ථ999

37 087

28 789

39 853

26 554

32 295

Coal

4 426

9 140

10 618

12 312

11 582

16 131

9 004

4 660

Gas

1 730

4 847

5 983

8 313

6 222

9 173

5 771

5 993

Oil

1 332

1 062

801

556

819

614

705

323

World

Nuclear

2 013

2 584

3 400

4 294

3 322

3 914

3 546

5 837

Hydro

2 144

3 490

4 555

5 827

4 412

5 478

4 667

6 394

173

992

2 642

5 785

2 432

4 544

2 861

9 089

Other renewables

© OECD/IEA, 2013

The evoluƟon of the power mix in OECD countries is markedly diīerent from that in nonOECD countries, with a stronger shiŌ towards low-carbon technologies, mainly renewables (Figureථ5.3). In OECD countries, coal-Įred generaƟon declines in absolute terms by almost one-quarter, compared to the level in 2011, and oil-Įred generaƟon by three-quarters. By contrast gas-Įred generaƟon grows as does nuclear generaƟon (to a lesser degree). Output from renewables sees the strongest growth, increasing by slightly more than the net growth in generaƟon in OECD countries, primarily led by the increase in wind power. In non-OECD countries, by contrast, coal remains the largest source of generaƟon by a wide margin with coal-Įred generaƟon meeƟng more than 30% of the growth of electricity demand, increasing the most of any source in absolute terms. However, generaƟon from all forms of renewables taken together increase even more in absolute terms and account for almost 40% of non-OECD incremental generaƟon from 2011-2035, led by hydropower and wind. In absolute terms the second-largest increase in non-OECD generaƟon from a singleenergy source comes from gas, primarily in the Middle East, China and India, due to the combinaƟon of burgeoning energy needs, the availability of gas and policies that support gas use in the power sector. Nuclear is the second-fastest growing source of generaƟon from a single energy source, behind non-hydro renewables. Chapter 5 | Power sector outlook

175

2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18

Figure 5.3 ‫ ٲ‬Electricity generation by source in the New Policies Scenario TWh

OECD

Non-OECD

10 000 8 000 6 000 4 000 2 000

1990

2035

2011 Coal

Renewables

1990 Gas

2011 Nuclear

2035 Oil

RelaƟve costs, which are partly inŇuenced by government policies, are one of the main drivers of the projected changes in the types of fuels and technologies used to generate power. For fossil-fuelled generaƟon, especially natural gas, the cost of generaƟon is very sensiƟve to fuel prices, while for nuclear power and renewables the capital cost of the plant is far more important. GeneraƟng costs for each technology vary widely across regions and countries, according to local fuel prices, regulaƟons and other cost factors (IEAͬNEA, 2010). Water scarcity, which can pose reliability risks for coal-Įred and nuclear plants that use large amounts of water for cooling, can inŇuence the generaƟon mix and generaƟng costs (IEA, 2012). In some regions, including parƟcular areas in Europe and the United States, public opposiƟon to building power infrastructure of almost all types – coal-Įred plants as well as wind turbines and transmission lines – is becoming a more important factor in determining the pace at which new projects can be completed.

ĂƉĂĐŝƚLJƌĞƟƌĞŵĞŶƚƐĂŶĚĂĚĚŝƟŽŶƐ In the New Policies Scenario, generaƟon capacity increases by almost three-quarters from 5ථ649 gigawaƩs (GW) in 2012 to 9ථ760ථGW in 2035 in order to saƟsfy growing demand needs and aŌer allowing for capacity closures over the period (Figureථ5.4).3 Over the projecƟon period, gross capacity addiƟons total 6ථ050ථGW, with about one-third replacing the reƟrement of 1ථ940ථGW (34% of current installed capacity). The majority of new plants are powered by gas (1ථ370ථGW), wind (1ථ250ථGW) and coal (1ථ180ථGW).

© OECD/IEA, 2013

In the Outlook, generaƟng capacity is reƟred once it reaches the end of its technical lifeƟme.4 The technical lifeƟmes of hydropower, coal and nuclear plants are longest (assumed to be 70 years for hydropower, 50ථyears for coal, and 40-60ථyears for nuclear depending on the country), followed by gas plants (40ථyears) and, wind turbines and solar PV installaƟons (20ථyears). Where the economic case is sound, owners may choose to invest in upgrades or 3.ഩ All electrical capacities presented in this Outlook are expressed in gross capacity terms. 4.ഩ Power plant lifetimes are expressed in both technical and economic terms. The technical lifetime corresponds to the design life of the plant. The economic lifetime is the time taken to recover the investment in the plant and is usually shorter than the technical lifetime (Tableථ5.5). 176

World Energy Outlook 2013 | Global Energy Trends

1

Figure 5.4 ‫ ٲ‬Installed capacity by source in the New Policies Scenario

3

GW

refurbishments to extend the lifeƟme of an ageing plant rather than to reƟre it and build a new one. In addiƟon, refurbishing an exisƟng power plant may oīer a short-term soluƟon to manage the risks from developing environmental regulaƟons, given the long lifeƟmes for new thermal power plants.

10 000 historical

projecons

Other renewables Wind

8 000

Nuclear Oil

4 000

6

Gas

7

2 000 Coal

1970

1980

1990

2000

2010

2020

8

2030 2035

Thermal power plants are older, on average, in OECD countries than in non-OECD countries, meaning that a higher proporƟon of OECD plants face reƟrement during the projecƟon period. At the end of 2012, almost two-thirds of the coal-Įred capacity in OECD countries was more than 30 years old (Figureථ5.5). By contrast, almost three-quarters of coal-Įred capacity in non-OECD countries is less than 20ථyears old. The majority of gas-Įred capacity in OECD countries, parƟcularly in the United States and European Union, is young as new plants in the last two decades have been largely gas (and renewables). Age proĮles for nuclear power capacity reŇect its earlier development in OECD countries and recent growth in developing countries, where most plants in operaƟon were built aŌer 1990.  JHSURÀOHRILQVWDOOHGWKHUPDOFDSDFLW\E\UHJLRQHQG Figure 5.5 ‫ ٲ‬$

10-20 years 20-30 years 30-40 years

Oil

40-50 years 50+ years

Gas

11 12

25%

25%

50%

75%

15 16

18

100%

Sources: PlaƩs World Electric Power Plants Database, December 2012 ediƟon͖ IAEA (2013).

Chapter 5 | Power sector outlook

14

17

Coal 50%

10

13

< 10 years Nuclear

75%

9

Non-OECD

OECD

100%

4 5

Hydro 6 000

© OECD/IEA, 2013

2

177

© OECD/IEA, 2013

178

Table 5.4 ‫& ٲ‬  XPXODWLYHFDSDFLW\UHWLUHPHQWVE\UHJLRQDQGVRXUFHLQWKH1HZ3ROLFLHV6FHQDULR (GW) Coal

Gas

Oil

OECD

265

178

147

81

Americas

109

104

59

98

96

123

Nuclear Bioenergy

GeoSolar PV thermal

Wind

60

80

231

6

10

19

36

86

46

10

16

22

31

41

45

34

33

42

46

26

10

35

42

195

183

92

CSP*

Marine

Total

124

3

0

1 176

3

13

1

0

440

74

3

12

1

-

377

35

135

1

81

2

0

530

7

10

10

1

31

0

0

205

25

5

7

5

1

27

-

-

157

103

36

22

20

167

4

36

0

0

765

113

17

32

1

1

5

0

2

-

-

262

43

80

5

20

1

-

0

0

0

-

-

149

78

17

25

2

13

9

150

3

31

0

0

329

China

43

1

3

-

7

3

122

0

26

0

0

205

India

26

3

2

1

3

3

26

-

3

-

-

68

0

30

38

-

0

1

1

-

0

-

-

69

22

13

10

-

0

2

2

0

1

-

-

50

3

10

14

1

7

7

10

1

1

-

-

54

2

1

1

1

5

5

8

-

1

-

-

24

World

460

361

249

117

82

100

398

10

160

3

0

1 941

European Union

130

33

43

42

34

27

134

1

81

2

0

528

United States Europe Asia Oceania Japan

World Energy Outlook 2013 | Global Energy Trends

Hydro

Non-OECD E. EuropeͬEurasia Russia Asia

Middle East Africa LaƟn America Brazil

*CSP с concentraƟng solar power.

© OECD/IEA, 2013

Chapter 5 | Power sector outlook

Table 5.5 ‫& ٲ‬  XPXODWLYHJURVVFDSDFLW\DGGLWLRQVE\UHJLRQDQGVRXUFHLQWKH1HZ3ROLFLHV6FHQDULR (GW) Coal

Gas

Oil

117

525

21

83

34

266

8

27

206

Europe

52

Asia Oceania

OECD Americas United States

Japan Non-OECD E. EuropeͬEurasia Russia

Nuclear Bioenergy

GeoSolar PV thermal

Hydro

Wind

113

147

611

24

23

49

60

231

6

19

40

32

156

5

31

49

32

102

8

29

9

77

7

1 065

850

84

CSP*

Marine

Total

367

24

13

2 046

12

104

13

2

802

169

8

92

11

1

611

66

325

4

165

9

9

870

15

21

55

8

98

3

3

374

3

10

15

27

4

79

-

1

232

63

219

134

593

635

18

385

45

1

4 007

177

1

51

10

29

21

3

7

-

0

384

38

116

0

33

7

18

6

2

1

-

-

222

902

353

11

150

94

370

534

11

302

18

1

2 745

China

454

142

1

114

57

188

384

2

177

13

0

1 533

India

288

100

2

26

16

82

106

0

92

4

0

717

1

153

31

7

4

12

26

-

32

14

-

281

70

87

11

5

10

61

18

2

26

11

-

302

8

79

9

5

17

120

36

2

17

3

-

295

4

32

5

3

12

72

29

-

9

1

-

166

1 182

1 374

84

302

247

740

1 246

42

752

70

14

6 052

European Union

49

129

5

29

47

48

311

3

162

9

9

800

Average economic lifeƟme (years)

30

25

25

35

25

50

20

25

20

20

20

7

6

Asia

Middle East Africa LaƟn America Brazil World

179

*CSP с concentraƟng solar power.

1

2

3

4

5

8

9

10

11

12

13

14

15

16

17

18

Future gross additions of baseload generation using coal, hydropower and nuclear are concentrated overwhelmingly in non-OECD countries. Gross additions of gas-powered plants during the projection period are also greater in non-OECD countries, which have installed a lower proportion of gas-fired capacity to date. More than 3ථ100ථGW of renewables is added worldwide, nearly double the present installed capacity, though these installations usually generate less electricity overall than the equivalent thermal counterpart, due to the lower capacity factors typically associated with variable resources (see Chapterථ6). In non-OECD countries, most capacity addiƟons are built to meet new demand. China installs more than 1ථ530ථGW during the projecƟon period (Figureථ5.6). Of this, 630ථGW is non-hydro renewables, accounƟng for over one-quarter of the global total in that category (about as much as the European Union and Japan combined). SigniĮcant addiƟons also come from coal (30% of the total), hydropower (12%) and gas (9%). China adds more nuclear capacity than the total installed nuclear capacity in the United States at present. Its total capacity addiƟons are more than twice those of India, where coal-Įred plant makes the biggest contribuƟon to gross capacity addiƟons (about 40%) over 2013-2035. These coal-Įred plants are split almost evenly between subcriƟcal and supercriƟcal technologies. Deployment of supercriƟcal technologies helps to raise the average eĸciency of Indian coal-Įred generaƟon from its very low present level. Non-hydro renewables account for some 30% of India’s addiƟons, boosted by the NaƟonal Solar Mission target and other support policies. Figure 5.6 ‫ ٲ‬Power generation gross capacity additions and retirements by VHOHFWHGUHJLRQLQWKH1HZ3ROLFLHV6FHQDULR Rerements United States

Capacity addions

European Union

Net capacity change

Japan China India Russia

© OECD/IEA, 2013

-600

-300

0

300

600

900

1 200

1 500

1 800 GW

In OECD countries, replacing reƟred capacity and decarbonising the power mix capacity are the main drivers of capacity addiƟons. In the European Union, signiĮcant reƟrements (530ථGW) and large-scale deployment of non-hydro renewables – which require greater 180

World Energy Outlook 2013 | Global Energy Trends

capacity addiƟons to ensure adequate system reliability – mean that the European Union sees the second-largest gross capacity addiƟons in the world during the projecƟon period (Figureථ5.7). Two-thirds of its addiƟons eīecƟvely replace capacity that is reƟred (including some wind and solar PV). The United States also installs signiĮcant capacity to replace reƟred units (62% of total gross addiƟons). One-third of addiƟons are gas, followed by wind (28%) and solar PV (15%). Limited coal capacity (4%) is added, with operators choosing to reƟre or refurbish exisƟng plants. Net addiƟons in Japan over 2013-2035 are relaƟvely low, but 54% of capacity operaƟng today is reƟred and has to be replaced (mainly gas and renewables).

Policies Scenario

4

GW

Renewables

7 8

500

2012 2035 United States

2012 2035 European Union

9

Addions

Rerements

1 000

Fossil fuels

Addions

Rerements

Addions

Rerements

Nuclear

2012

10 11

2035 China

Around 20% of the global gross addiƟons of thermal capacity projected through 2035, or about 590ථGW, are already under construcƟon. Nearly all of this capacity will be in operaƟon by 2018, though some nuclear reactors will come online later. Of the thermal capacity being built at present, 56% is coal-Įred and 28% is gas-Įred. These Įgures may understate the role that is likely to be played in the medium term by gas, as gas-Įred plants can be built much faster than coal-Įred plants: combined-cycle gas turbines (CCGTs) can usually be built within 2-3 years and open-cycle gas turbines in 1-2ථyears, while coal-Įred power plants oŌen take more than four years to start generaƟng electricity.

The amount of generaƟon from fossil fuels (coal, gas and oil) in each of the three scenarios depends, among other things, on policy factors such as the implementaƟon and level of carbon prices, the strength of support for renewables and nuclear, and the stringency of environmental regulaƟons. In the New Policies Scenario, the share of fossil fuels in total generaƟon falls from 68% to 57%.

Chapter 5 | Power sector outlook

12 13 14 15 16

&ŽƐƐŝůͲĨƵĞůůĞĚŐĞŶĞƌĂƟŽŶ

© OECD/IEA, 2013

3

6

2 500

1 500

2

5

Figure 5.7 ‫ ٲ‬Power capacity changes in selected regions in the New

2 000

1

181

17 18

In this scenario, global coal-Įred generaƟon increases from 9ථ140ථTWh in 2011 to 12ථ310ථTWh in 2035 (or by 35%), despite its share of total generaƟon falling from 41% to 33%. All of the growth comes from non-OECD countries, which are projected to conƟnue to rely on coal as a secure and aīordable means to support economic growth and development. In China, generaƟon from coal grows by half during the projecƟon period, though the average rate of growth slows from 2.2% per year unƟl 2020 to 1.2% per year thereaŌer (Figureථ5.8). Notwithstanding strong eīorts to diversify its power sector, China’s coal-Įred generaƟon in 2035 is projected to exceed present generaƟon from all sources in the United States and Japan combined, at more than 5ථ500ථTWh. India’s coal-Įred generaƟon more than doubles, making it the second-largest user of coal in the power sector by the end of the projecƟon period. In OECD countries, the improving compeƟƟveness of alternaƟves, resƟng on policies that increasingly promote low-carbon sources of generaƟon, lead to a 23% decline in coal-Įred generaƟon over 2011-2035. Figure 5.8 ‫& ٲ‬  RDOÀUHGSRZHUJHQHUDWLRQE\UHJLRQLQWKH1HZ3ROLFLHV Scenario TWh

14 000 12 000

India

10 000 8 000

China

6 000 European Union United States

4 000 2 000

1990

Rest of world 1995

2000

2005

2010

2015

2020

2025

2030

2035

© OECD/IEA, 2013

The average eĸciency of coal-Įred generaƟon worldwide improves from 36% to 40% during the projecƟon period as old plants, based on subcriƟcal technology, are reƟred and are increasingly replaced by supercriƟcal and other higher eĸciency technologies, such as ultra-supercriƟcal, integrated gasiĮcaƟon combined-cycle (IGCC) and combined heat and power (CHP) plants. ContribuƟng to the shiŌ in technologies are increases in carbon pricing (see Chapterථ1) and the fuel savings resulƟng from higher eĸciency, which result in lower fuel costs, and can reduce import dependency. The development of carbon capture and storage (CCS) remains limited in the New Policies Scenario, with 56ථGW of coal-Įred power plants ĮƩed with CCS generaƟng about 390ථTWh in 2035, around 3% of total coal-Įred power generaƟon. For each region, the average level of eĸciency of coal-Įred generaƟon reached at the end of the projecƟon period depends on the extent of reƟrement of old subcriƟcal plants, the rate of construcƟon of new plants, the type of technologies chosen for new plants and the 182

World Energy Outlook 2013 | Global Energy Trends

way that the coal-Įred Ňeet is operated. In China, the share of generaƟon from supercriƟcal and high eĸciency coal capacity rises from one-third to two-thirds over 2011-2035, raising average eĸciency from 36% to 40% (Figureථ5.9). The present average eĸciency of India’s coal-Įred generaƟon is extremely low because of heavy reliance on an ageing subcriƟcal Ňeet and the use of low quality coal. The projected addiƟon of new plants, some of which use subcriƟcal and supercriƟcal technologies, improves average eĸciency by eight percentage points, from 28% to 36%. In the European Union the eĸciency of coalĮred generaƟon increases from 38% to 44% as subcriƟcal plants are almost enƟrely phased out of service by 2035. In the United States, liƩle coal-Įred capacity is added during the projecƟon period (limiƟng opportuniƟes to deploy high eĸciency technologies) and many exisƟng plants are refurbished to extend their lifeƟme, resulƟng in only a small gain in the overall eĸciency of coal-Įred generaƟon.

1 2 3 4 5 6

Figure 5.9 ‫ ٲ‬6 KDUHRIFRDOÀUHGSRZHUJHQHUDWLRQE\WHFKQRORJ\DQGDYHUDJH

7

HIÀFLHQF\LQVHOHFWHGUHJLRQVLQWKH1HZ3ROLFLHV6FHQDULR

80%

45%

60%

40%

40%

35%

10

20%

30%

11

25%

12

© OECD/IEA, 2013

2011 2035 United States

2011 2035 European Union

2011 2035 China

2011 2035 India

Efficiency

50%

Share of coal generaon

100%

High efficiency

8

Supercrical Subcrical Average efficiency (right axis)

Gas-Įred generaƟon rises from 4ථ847ථTWh in 2011 to 8ථ310ථTWh in 2035 (or by 72%), its share of total generaƟon remaining constant at 22%. Gas-Įred generaƟon ĮƩed with CCS accounts for less than 1% of total gas-Įred generaƟon. Nearly 80% of incremental generaƟon over 2011-2035 comes from non-OECD countries. Around 20% of incremental generaƟon comes from the Middle East. In China, driven by the policy of increasing gas use to diversify the energy mix (Figureථ5.10), gas generaƟon is expected to increase eight-fold by 2035, to reach an absolute level slightly exceeding that of the European Union today. In the United States, the availability of relaƟvely cheap gas throughout the projecƟon period – the combined result of booming shale gas producƟon and a compeƟƟve gas market – underpins a 38% expansion of gas-Įred generaƟon. For countries in the European Union, the combinaƟon of low electricity demand growth, support to renewables, high gas-tocoal price spreads and low CO2 prices sƟŇes addiƟonal gas-Įred generaƟon in the period to 2020͖ beyond that, gas-Įred generaƟon increases as ineĸcient coal capacity is reƟred,

Chapter 5 | Power sector outlook

183

9

13 14 15 16 17 18

CO2 prices rise and the need for system Ňexibility becomes greater (to complement the large-scale deployment of renewables). Despite the higher gas prices in the European and Asia-PaciĮc markets, gas-Įred generaƟon sƟll has characterisƟcs that make it an aƩracƟve opƟon relaƟve to the alternaƟves, notably, lower capital costs, shorter construcƟon Ɵmes, greater operaƟonal Ňexibility and lower emissions. Figure 5.10 ‫* ٲ‬  DVÀUHGSRZHUJHQHUDWLRQE\VHOHFWHGUHJLRQLQWKH1HZ TWh

Policies Scenario 2 000

2011

1 600

Incremental generaon, 2011-2035

1 200 China

800

400

Russia

European Union

Middle East

United States

Developing Asia

Box 5.1 ‫& ٲ‬  RDOWRJDVVZLWFKLQJLQWKHSRZHUVHFWRU In power systems which have suĸcient spare capacity (e.g. in the United States), compeƟƟon between coal-Įred plants and combined-cycle gas turbines (CCGTs) can result in fuel switching between coal and gas. Dispatch choices depend on which of the plants can be run at lower cost and, consequently, on plant eĸciencies and relaƟve fuel prices. Coal and gas prices vary across regions and Ňuctuate over Ɵme, so the potenƟal for fuel-switching also varies. In some cases, as in the European Union, relaƟve fuel prices are aīected by CO2 prices that beneĮt gas-Įred generaƟon because of its lower carbon intensity relaƟve to coal.

© OECD/IEA, 2013

Fuel switching results from the changing circumstances of relaƟve prices in the underlying coal and gas markets. Sustained coal-to-gas switching requires either a substanƟal expansion of gas supply, leading to falling gas prices, (shale gas development in China, for example), stronger carbon pricing or environmental regulaƟons, or gas pricing reforms that encourage gas pricing based on gas-to-gas compeƟƟon, rather than oil-indexaƟon. Such fuel switching is unlikely to happen in rapidly growing and Ɵght power systems in Asia or Africa. The rapid growth in shale gas producƟon over the last several years has caused a dramaƟc fall in natural gas prices in the United States. They averaged $2.7ථper million BriƟsh thermal units (MBtu) in 2012, leading to an unprecedented switch from coal to

184

World Energy Outlook 2013 | Global Energy Trends

gas in the US power sector (and a signiĮcant drop in CO2 emissions). But as of mid2013, gas prices had risen to around $3.7ͬMBtu, allowing coal to regain some market share. Without environmental or other regulaƟons to set a price on CO2 emissions, the bulk of the coal-Įred Ňeet becomes compeƟƟve with CCGTs at current eĸciencies when gas prices are in the range of $4.5-5ͬMBtu.

1 2 3 4

Figure 5.11 ‫ ٲ‬Electricity generating costs for coal and gas by selected

Dollars per MWh (2012)

UHJLRQDQGIRUIXHOSULFHV 120

Variable cost range, 2008-2012:

100

Coal-fired generaon

6

80

Gas-fired generaon

7

60

2012 average

8

40

9

20

United States

European Union

10

Japan

Notes: MWh с megawaƩ-hour. The assumed coal-Įred plant eĸciency is 40%͖ gas-Įred plant eĸciency is 57%. Fuel prices are assumed to be the spot prices for central Appalachian coal and Henry Hub gas in the United States͖ the spot prices for ARA coal and NBP gas in Europe͖ the MCR Japanese marker for coal and LNG import price for gas in Japan. GeneraƟng costs in Europe include CO2 prices.

© OECD/IEA, 2013

5

In Europe, fuel switching depends on the interplay of prices for coal, gas and emiƩed CO2. The combinaƟon of soŌening CO2 prices, low coal prices (the result of ample coal availability on the internaƟonal market) and high gas prices has recently favoured coalĮred generaƟon in Europe, leading to a gas-to-coal switch. Taking the coal, gas and CO2 prices in the New Policies Scenario, coal looks set to remain economically preferable to gas in Europe through 2020. For the CO2 price to alter the outlook, it would have to be much higher than it is at present. For example, in 2020 a highly eĸcient CCGT would require a CO2 price of more than $60ͬtonne to displace a typical coal-Įred plant with an eĸciency of 39%. Fuel switching potenƟal in the Asia-PaciĮc region is very limited, due to very high prices for imported liqueĮed natural gas. In Japan and
Chapter 5 | Power sector outlook

185

11 12 13 14 15 16 17 18

Oil-Įred generaƟon is projected to decline from 1ථ062ථTWh in 2011 to just below 560ථTWh in 2035 (or by 48%), conƟnuing its long-term historical decline, its share of total generaƟon falling from 5% to 1.5% during the projecƟon period. In the OECD countries, it falls to a mere 0.6%. High oil prices, cuts to expensive fuel price subsidies in some countries and the falling relaƟve costs of alternaƟves make the economics of oil-Įred generaƟon increasingly unaƩracƟve. In most regions, oil is consigned to only a marginal role as emergency backup and in distributed applicaƟons in remote areas, or is used where gas distribuƟon networks are under-developed. The use of oil for power generaƟon falls in almost all regions. The decline is slowest in Africa and in the Middle East, the laƩer accounƟng for almost half of global oil-Įred generaƟon in 2035, because of the assumed persistence of fuel price subsidies in several countries and strong electricity demand growth.

EƵĐůĞĂƌƉŽǁĞƌ5 There were 437 nuclear reactors in operaƟon worldwide at the end of 2012, with a capacity of 394ථGW (IAEA, 2013).6 More than 80% of capacity is in OECD countries, 11% in Eastern Europe and Eurasia, and 8% in developing countries. Though their share of installed capacity is low today, non-OECD countries will account for the bulk of future growth. Of the 73ථGW presently under construcƟon, about 80% is in non-OECD countries.

© OECD/IEA, 2013

In the United States, lower electricity prices (as a result of cheap gas) and high repair costs have led to the reƟrement of four reactors at three power plants in 2012-2013. ConstrucƟon has begun on two new units and a further two units have received construcƟon licences. In
5.ഩ The WEO-2014 will feature an in-depth analysis of the outlook for nuclear power. 6.ഩ In net terms (excluding the own use of electricity within the plants) this is equivalent to 373ථGW. 186

World Energy Outlook 2013 | Global Energy Trends

In the New Policies Scenario global nuclear generaƟon grows from 2ථ584ථTWh in 2011 to 4ථ300ථTWh in 2035, its share of total generaƟon remaining constant at 12%. Growth in generaƟon is underpinned by a corresponding expansion of capacity, which rises from 394ථGW in 2012 to 578ථGW in 2035. This is the net result of 117ථGW of reƟrements and 302ථGW of capacity addiƟons during the projecƟon period. The rate of expansion of nuclear power conƟnues to be mainly policy driven. It expands in markets where there is a supporƟve policy framework, which in some cases acƟvely targets a larger role for nuclear in the mix in order to achieve energy security aims. But policy frameworks can also hinder or eliminate nuclear power, oŌen as a result of public opposiƟon: even where there is no explicit ban, long permiƫng processes, such as in the United States, can signiĮcantly hinder development by increasing uncertainty about project compleƟon and increasing costs.

1 2 3 4 5 6

Figure 5.12 ‫ ٲ‬Nuclear power installed capacity by region in the New

GW

Policies Scenario

7

350 OECD 300 OECD excluding Korea

8

China

9

250 200 150 100 50

© OECD/IEA, 2013

1970

1980

1990

2000

2010

2020

Other non-OECD

10

Eastern Europe / Eurasia

11

2030 2035

12

The largest nuclear gross capacity addiƟons are in China, which adds 114ථGW during the projecƟon period (or 38% of global nuclear addiƟons before taking into account reƟrements). Of the total projected to be added, 28% is already under construcƟon, with building expected to begin on several other plants by 2015. Russia adds 33ථGW, the second-largest total globally (though around 60% is needed to replace units that are reƟred). Among OECD countries,
Chapter 5 | Power sector outlook

187

13 14 15 16 17 18

ZĞŶĞǁĂďůĞƐ7 GeneraƟon from renewable energy sources conƟnues to increase rapidly, growing more than two-and-a-half Ɵmes over the projecƟon period, from 4ථ482ථTWh to over 11ථ600ථTWh and accounƟng for almost half of total incremental generaƟon over the period. This growth is driven by improving compeƟƟveness, a result of falling costs for renewables technologies, rising fossil fuel prices and carbon pricing, but mainly government support, in the form of subsidies to accelerate the deployment of renewables (see Chapterථ6). The share of renewables in the overall mix grows from 20% to 31% during the projecƟon period. Hydropower remains the largest source of renewables generaƟon, conƟnuing to provide 16% of total generaƟon over the projecƟon period. Similar to the absolute level of growth in hydropower, wind generaƟon grows by some 2ථ340ථTWh, the third-largest increment behind only gas and coal. GeneraƟon from solar PV increases at a much higher rate than wind, reaching 950ථTWh in 2035. Figure 5.13 ‫ ٲ‬5  HQHZDEOHVEDVHGSRZHUJHQHUDWLRQDQGVKDUHRIWRWDO generation by region in the New Policies Scenario 90%

Other

3 000

75%

Geothermal

2 400

60%

Biomass

1 800

45%

Wind

1 200

30%

600

15%

TWh

3 600

Solar PV

Hydro Share of total generaon (right axis)

© OECD/IEA, 2013

2011 2035 2011 2035 2011 2035 2011 2035 2011 2035 2011 2035 United European China India Other Other States Non-OECD Union OECD

Two-thirds of incremental growth in renewables generaƟon occurs in non-OECD countries. China, which is already the world’s largest producer of renewable electricity, accounts for 28% of global growth (more than the combined growth of the European Union, the United States and Japan), its renewables generaƟon more than tripling from 814ථTWh in 2011 to 2ථ800ථTWh in 2035 (Figureථ5.13). The United States and the European Union both see generaƟon from renewable sources double. Hydropower plays a much more signiĮcant role in some regions than in others. It accounts for 44% of incremental renewables generaƟon in non-OECD countries, where a sizeable amount of cost-compeƟƟve potenƟal is untapped at present. OECD countries, by contrast, have already developed much of their economic hydropower potenƟal and incremental renewables generaƟon comes mainly from wind (47%), biomass (16%) and solar PV (16%). 7.ഩ A more detailed analysis of the prospects for renewables for heat and power generation can be found in Chapterථ6. 188

World Energy Outlook 2013 | Global Energy Trends

dƌĂŶƐŵŝƐƐŝŽŶĂŶĚĚŝƐƚƌŝďƵƟŽŶ Transmission and distribuƟon (TΘD) networks are criƟcal to deliver electricity reliably to end-users. Reinforcement and expansion of TΘD networks will be necessary to maintain or improve the quality of service to exisƟng customers, provide access for new end-users (mainly in developing countries) and connect new sources of generaƟon. BoƩlenecks exist today in many TΘD networks around the world and removing them would help reduce losses and improve the access to available spare capacity that is needed in power systems. However, like other energy infrastructure, TΘD projects in several regions, parƟcularly in OECD countries, face public opposiƟon that delays projects. In the New Policies Scenario, the length of TΘD lines globally expands from some 69ථmillion kilometres (km) in 2012 to 94ථmillion km in 2035. DistribuƟon networks deliver power over short distances from substaƟons to end-users, whereas transmission grids transport power over long distances from generators to local substaƟons near customers. Because of their much higher density, distribuƟon networks account for more than 90% of the total length of TΘD networks at present and more than 85% of growth during the projecƟon period. TΘD networks increase most in China (7ථmillionථkm) and India (3.5ථmillionථkm), to cover increasing demand and the connecƟon of new end-users (Figureථ5.14). By 2035, around 50% of today’s grid infrastructure will have reached 40ථyears of age, which is the average technical lifeƟme of TΘD assets, highlighƟng the need for signiĮcant investment in refurbishments and replacements during the projecƟon period.8 The age of the grid varies across regions: younger grids can be found in regions with recent and ongoing infrastructure expansions (such as China, Southeast Asia and Africa), while in Europe, the United States and Russia current grid infrastructure is older and 60% or more need to be refurbished or replaced over the Outlook period. Figure 5.14 ‫ ٲ‬Existing and additional kilometres of transmission and distribution lines by selected region in the New Policies Scenario

United States

Distribuon

India

3 4 5 6 7 8 9 10 11 12

14

Year: 2012

European Union

2

13

Type: Transmission

China

1

Change to 2035

Brazil

15 16

Middle East Russia

© OECD/IEA, 2013

4

8

12

16

20 24 Million km

17

8.ഩ The average technical lifetimes of TΘD assets vary according to the type of asset, the conditions under which they are used and maintenance performed. In some cases, they operate much longer than 40 years.

Chapter 5 | Power sector outlook

189

18

TΘD networks play a fundamental part in enhancing system Ňexibility and in providing for increased use of renewables. The best sites for some renewable energy technologies are located far from demand centres and addiƟonal high-voltage transmission lines are needed to exploit them. The growing contribuƟon of power generaƟon from renewables necessitates increased co-ordinaƟon between grid infrastructure and renewable projects (see Chapterථ6). Future grids are expected increasingly to deploy smart grid technologies, such as digital communicaƟon and control technologies, to co-ordinate the needs and capabiliƟes of electricity generators, end-users and grid operators. AddiƟonal beneĮts include greater system reliability, a lower cost of electricity supply (through fuel savings and avoided investment in addiƟonal generaƟon capacity) and reduced environmental impact.

CO2ĞŵŝƐƐŝŽŶƐ Increasing penetraƟon of low-carbon technologies and improvements in the thermal eĸciency of fossil-fuelled power plants help to temper the growth in CO2 emissions from the power sector. In the New Policies Scenario, CO2 emissions from the global power sector rise from 13.0ථgigatonnes (Gt) in 2011 to 15.2ථGt in 2035, including some 1.3ථGt from heat producƟon throughout the period. Emissions growth slows with Ɵme, falling from 0.9% per year over 2011-2020 to 0.5% per year during the remaining period, while electricity generaƟon grows by 2.7% per year and 1.9% per year, over the respecƟve periods.

© OECD/IEA, 2013

Globally, the CO2 emissions intensity of electricity generaƟon falls from 532ථgrammes of CO2 per kilowaƩ-hour (gථCO2ͬkWh) in 2011 to 374ථgථCO2ͬkWh in 2035 (an improvement of 30%). In the Current Policies Scenario, emissions from the world power sector total 19.1ථGt in 2035. Increased use of low-carbon technologies (including a small amount of CCS) is responsible for 42% of the savings in the New Policies Scenario, compared with the Current Policies Scenario, reduced energy demand is responsible for 53% and coal-to-gas switching and improvements in the average eĸciency of coal- and gas-Įred power plants are responsible for the rest. A shiŌ toward higher eĸciency coal-Įred plants lowers the CO2 emissions intensity of India’s power sector by one-third, though a signiĮcant expansion of generaƟon causes emissions to double during the projecƟon period (Figureථ5.15). China sees its CO2 emissions intensity decline by 36% for similar reasons (in addiƟon to a strong eīort to diversify the mix towards low-carbon sources) while emissions from electricity generaƟon increase in total by onethird. OECD power sector CO2 emissions fall throughout the projecƟon period, as limited growth in electricity demand is provided by a power mix with a steeply declining emissions intensity. Today, the emissions intensity of the European Union (345ථgථCO2ͬkWh) is slightly lower than the average of a CCGT plant9 (355ථgථCO2ͬkWh), but by 2035 it is around 45% of the present average (160ථgථCO2ͬkWh), reŇecƟng the large increment of generaƟon that comes from renewables over 2011-2035. 9.ഩ With an assumed 57% efficiency. 190

World Energy Outlook 2013 | Global Energy Trends

Figure 5.15 ‫ ٲ‬CO2 emissions intensity in the power sector and electricity

1

generation by region in the New Policies Scenario

CO2 emissions intensity (g CO2/kWh)

(a) 2011 1 000

2

India 0.9 Gt

800

3

600

4 China 3.6 Gt

400

United States 2.2 Gt

200

5

Other non-OECD 2.6 Gt

Other EU OECD 1.4 Gt 1.1 Gt

6

0 0

3

9

6

12

15

18

21

24

7

27 30 33 36 Generaon (thousand TWh)

8

CO2 emissions intensity (g CO2/kWh)

(b) 2035 1 000

9

800

10

600

11

400 India 1.9 Gt

200

China 4.9 Gt

United States 1.9 Gt

Other non-OECD 3.6 Gt

0 0

3

6

9

12

15

18

21

24

Other OECD 1.1 Gt

12

EU 0.6 Gt

13

27 30 33 36 Generaon (thousand TWh)

Note: EU с European Union.

Investment

14

10

© OECD/IEA, 2013

SubstanƟal investments in the power sector will be required over the projecƟon period to saƟsfy rising electricity demand and to replace or refurbish ageing infrastructure. In the New Policies Scenario, cumulaƟve global investment in the power sector is $17.0ථtrillion (in year-2012 dollars) over 2013-2035, averaging $740ථbillion per year (Figureථ5.16).11 New generaƟng capacity accounts for 58% of total investment, while the remainder is needed 10.ഩA WEO special report analysing the investment and financing needs of the world’s energy infrastructure will be published in mid-2014. 11.ഩ Investment assumptions can be found at www.worldenergyoutlook.org/weomodel/investmentcosts/.

Chapter 5 | Power sector outlook

191

15 16 17 18

in TΘD networks. Renewables account for 62% of investment in new power plants, led by wind, hydropower and solar PV. The share of investment in renewables is higher than their share of capacity addiƟons (just over half), reŇecƟng their higher capital costs compared with fossil-fuelled capacity. Average annual investment in renewables rises marginally to 2025, before picking up at a faster rate due to an escalated rate of deployment at the end of the projecƟon period (which is partly due to the reƟrement of renewables capacity). ReducƟons in the costs of renewable technologies parƟally oīset the eīect of elevated deployment towards the end of the projecƟon period. Figure 5.16 ‫ ٲ‬Power sector cumulative investment by type and region in the 1HZ3ROLFLHV6FHQDULR China Transmission and distribuon

India Other non-OECD

Renewables

United States European Union Other OECD

Fossil fuels

Nuclear 1 000 2 000

3 000 4 000

5 000 6 000 7 000 8 000 Billion dollars (2012)

CumulaƟve investment in TΘD infrastructure projected over 2013-2035 in the New Policies Scenario is $7.1ථtrillion, or about 42% of total power sector investment. Twothirds of the cumulaƟve investment takes place in non-OECD countries, where strong growth in electricity demand necessitates the construcƟon of new TΘD lines. While 68% of TΘD investment in non-OECD countries goes to the installaƟon of new lines, 29% goes to the refurbishment and replacement of exisƟng lines, with the remaining investment needed for supporƟng the integraƟon of increasing renewables capacity. China accounts for one-quarter of the investment in TΘD infrastructure worldwide. In OECD countries, refurbishment and replacement of exisƟng assets accounts for the bulk of TΘD investments, while one-third goes to build new lines that saƟsfy demand growth. This is due to the age structure of the assets, but also reŇects relaƟvely stable energy demand. Increasing renewables deployment means that 5% goes to renewables integraƟon.

© OECD/IEA, 2013

Electricity prices End-user electricity prices are determined by the underlying costs of supplying electricity – including the cost of generaƟng electricity, transmiƫng and distribuƟng it through the network, and selling it to the Įnal customer – and by any taxes or subsidies applied by governments to electricity sales. In many countries, the costs of subsidies to renewable energy are also passed on to the consumers through the electricity price. 192

World Energy Outlook 2013 | Global Energy Trends

Diīerences in wholesale electricity prices are a primary driver of diīerences in enduser electricity prices between regions, although subsidies, taxes, grid costs and support mechanisms can have a signiĮcant inŇuence. In the United States, wholesale prices are projected to be among the lowest in the world, having fallen in recent years. This expectaƟon stems mainly from cheaper gas from abundant domesƟc shale gas supplies, which reduce fuel costs and investment costs, as CCGT plants have one of the lowest capital costs. Wholesale prices in the European Union are projected to be 75% higher than in the United States in 2035. Strong deployment of wind and solar PV lowers fuel costs in the European Union, but raises operaƟon and maintenance (OΘM) costs and investment costs. Slowly rising gas prices – gas-Įred generaƟon maintaining a share of around 20% of the mix throughout the projecƟon period – and increasing CO2 costs also drive up European Union wholesale prices over Ɵme. Japan’s power system has been under extreme stress following the accident at Fukushima Daiichi and the subsequent reducƟon in generaƟon from nuclear power plants. Japan’s fossil-fuelled power plants have had to run much more frequently to meet electricity demand (even with strong eīorts to reduce demand). For example, the Japanese Ňeet of oil-Įred power plants ran at an average capacity factor of 20% in 2010, but over 40% in 2012. As Japan relies heavily on high-cost imported fossil fuels, total fuel costs for power generaƟon have risen substanƟally in recent years. With some of its nuclear power plants expected to come back online and increased generaƟon from renewable sources (wind, solar and geothermal), expensive oil- and gas-Įred generaƟon is projected to fall, lowering fuel import bills and reducing wholesale prices, improving compeƟƟveness by the end of the projecƟon period. However, wholesale prices in Japan are sƟll more than 90% higher than in the United States in 2035.

© OECD/IEA, 2013

China’s wholesale prices are also among the lowest in the world, though they increase to 23% above the level of the United States in 2035. In large part, they are pushed up by the assumed CO2 price, though whether or not the associated costs will be passed on to consumers depends on the eventual design of the market. Total CO2 costs for power generaƟon in China reach nearly $150ථbillion at the end of the projecƟon period, Įve Ɵmes its level of support for renewables in the same year. ConƟnued expansion of generaƟon from nuclear, renewables and coal helps to keep the fuel cost per unit of generaƟon broadly constant, despite upward pressure from expanded gas-Įred generaƟon, but raises investment costs per unit of generaƟon. Wholesale prices can vary markedly within countries or regions, such as in the United States and Europe, depending on the design and characterisƟcs of intra-regional electricity markets. Moreover, they can Ňuctuate from year-to-year with changes in the weather, economic condiƟons, fuel prices and unexpected events. VolaƟlity in wholesale prices can complicate the Įnancing and building of new power plants, parƟcularly capital-intensive projects such as nuclear power plants, potenƟally limiƟng installed capacity and puƫng upward pressure on prices. The projecƟons of wholesale prices ensure that all plants in operaƟon recover the direct costs of generaƟng electricity (the variable costs of generaƟon),

Chapter 5 | Power sector outlook

193

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18

and that newly built power plants are able to recover all of their Įxed costs in addiƟon to their variable costs. Based on gradual fuel price changes, electricity prices in the projecƟons tend to evolve smoothly over Ɵme. However, real-world prices will vary around these longterm trends, due to short-term fuel price volaƟlity and investment cycles.

ZĞƐŝĚĞŶƟĂů ResidenƟal electricity prices are projected to increase in nearly all regions through 2035 along with generally rising fuel prices worldwide. In addiƟon to the wholesale cost, residenƟal prices (excluding taxes) take account of transmission and distribuƟon network costs, retailing and, where appropriate, the cost of renewable energy subsidies passed on to consumers. Wholesale prices remain low, helping to keep residenƟal prices in the United States amongst the lowest in OECD countries (Figureථ5.17). ResidenƟal prices in the European Union stabilise around 2030, despite conƟnually rising wholesale prices. This occurs largely due to the improving compeƟƟveness of renewables and the expiraƟon of subsidy commitments to higher cost renewables, which receive decreasing subsidies per unit of generaƟon. Increased network costs stem principally from refurbishment and extension of grid infrastructure, with a smaller part from renewables integraƟon. Within the European Union, current residenƟal end-user prices (excluding taxes) span a wide range, with prices in 2012 as high as $240 per megawaƩ-hour (MWh) in Ireland and as low as $120ͬMWh in France, and this is expected to conƟnue. In Japan, residenƟal prices follow the wholesale price trend, declining conƟnually aŌer 2012 as the power system stabilises. In China, rising wholesale prices contribute to the more than 30% increase in residenƟal prices in real terms between 2012 and 2035. By 2035, residenƟal prices in China are sƟll about one-third below the level of the United States. Figure 5.17 ‫ ٲ‬Average residential electricity prices (excluding taxes) by

Dollars per MWh (2012)

region and cost component in the New Policies Scenario 280

Renewables subsidy

240

Network, retail and other

200

Wholesale price component: CO2 price Fuel O&M Investment costs and net revenues

160 120 80

© OECD/IEA, 2013

United States

European Union

Japan

2035

2020

2030

2012

2035

2030

2020

2012

2035

2030

2012

2020

2035

2030

2020

2012

40

China

Notes: Hatched areas represent subsidies that are partly or fully borne by taxpayers rather than consumers. Chinese prices have a low component to cover network, retail and other costs, due to subsidisaƟon.

194

World Energy Outlook 2013 | Global Energy Trends

ResidenƟal electricity bills are projected to increase as a result of the combinaƟon of higher prices and, in most regions, higher per-capita electricity consumpƟon. Per-capita spending on electricity in the New Policies Scenario rises by 26% in the European Union, to around $375 in 2035͖ by 14% in the United States, to almost $600͖ and by only 2% in Japan, to about $640. The share of income spent on electricity falls during the projecƟon period, as incomes increase more quickly than electricity bills in these OECD countries. In China, percapita spending on electricity triples, reaching $120 in 2035, as electricity prices increase and per-capita electricity consumpƟon rises sharply. Despite this rapid increase, income per capita more than keeps up, reducing the share of income spent on electricity over Ɵme.

Projected electricity prices to industrial consumers in the New Policies Scenario show a modest widening of diīerenƟals between the United States and regions such as the European Union and China over 2012-2035 (Figureථ5.18). This is partly explained by larger increases of wholesale costs in the European Union and China. European Union industry prices increase by 24% during the projecƟon period and, by 2035, are the highest in major industrialised countries and roughly twice the level of those in the United States. Along with a decline in the cost of renewables subsidies, the European Union experiences rising network, retail and other costs in the laƩer half of the Outlook period, in part due to the rapid deployment of renewables. Industrial prices in Japan are extremely high at present, given the extent of dependency on fossil fuel imports, but these fall over Ɵme to levels similar to those in the European Union. China’s industrial prices rise by almost 20%, remaining much higher than prices in the United States through 2035. The compeƟƟve implicaƟons of these diīerences in prices between regions are discussed in Chapterථ8. Figure 5.18 ‫ ٲ‬Average industry electricity prices (excluding taxes) by region and cost component in the New Policies Scenario Dollars per MWh (2012)

200

Renewables subsidy

160

Network, retail and other

120

Wholesale price component: CO2 price Fuel O&M Investment costs and net revenues

80

United States

European Union

Japan

3 4

7 8 9 10 11 12 13 14 15 16 17

China

Notes: Hatched areas represent subsidies that are partly or fully borne by taxpayers rather than consumers. The prices presented exclude taxes (see Chapter 8, Figure 8.7 for prices inclusive of taxes). Industry and residenƟal wholesale prices are assumed to be equal but could diīer, for example, due to long-term power purchasing agreements.

Chapter 5 | Power sector outlook

6

2035

2030

2012

2020

2035

2030

2012

2020

2035

2030

2012

2020

2035

2030

2020

40

2012

2

5

/ŶĚƵƐƚƌLJ

© OECD/IEA, 2013

1

195

18

© OECD/IEA, 2013

Chapter 6 Renewable energy outlook Basking in the sun? Highlights

x The share of renewables in primary energy use in the New Policies Scenario rises to 18% in 2035, from 13% in 2011, resulƟng from rapidly increasing demand for modern renewables to generate power, produce heat and make transport fuels. LimiƟng this rapid growth is the conƟnued shiŌ away from the use of tradiƟonal biomass in developing countries in favour of modern energy services.

x Power generaƟon from renewables increases by over 7ථ000ථTWh from 2011 to 2035, making up almost half of the increase in total generaƟon. Renewables become the second-largest source of electricity before 2015 and approach coal as the primary source by 2035, with conƟnued growth of hydropower and bioenergy, plus rapid expansion of wind and solar PV. Almost two-thirds of the increase in power generaƟon from renewables is in non-OECD countries. The increase in China is more than that in the European Union, United States and Japan combined.

x ConsumpƟon of biofuels increases from 1.3ථmboeͬd in 2011 to 4.1ථmboeͬd in 2035, to meet 8% of road-transport fuel demand in 2035. The United States, Brazil, European Union and China make up more than 80% of all biofuels demand. Advanced biofuels, helping to address sustainability concerns about convenƟonal biofuels, gain market share aŌer 2020, reaching 20% of biofuels supply in 2035.

x CumulaƟve investment of $6.5ථtrillion is required in renewable energy technologies from 2013 to 2035, only 5% of which is for biofuels. Renewables account for 62% of investment in new power plants through to 2035. In addiƟon, investments in new transmission and distribuƟon lines of $260 billion are needed for the integraƟon of renewables. Increasing generaƟon from wind and solar PV has impacts on power markets and system operaƟon, which can reduce the proĮtability of other generators, but also sƟmulate changes in market design.

x Renewable energy technologies are becoming more compeƟƟve compared to wholesale electricity prices, but their conƟnued growth hinges on subsidies to facilitate deployment and drive further cost reducƟons. Subsidies to renewables reached $101ථbillion in 2012, up 11% relaƟve to 2011. Almost 60% of these were paid in the European Union. Global subsidies to renewables increase to over $220ථbillion by 2035. Wind becomes compeƟƟve in a growing number of regions, as does solar PV, but only in a limited number of markets.

© OECD/IEA, 2013

x Along with reducing CO2 emissions, deploying renewables delivers co-beneĮts, including reducƟon of other pollutants, enhancing energy security, lowering fossil-fuel import bills and fostering economic development. The challenge is to design creaƟve renewable support schemes that are eīecƟve and cost-eĸcient, but also take into consideraƟon exisƟng and planned infrastructure in order to minimise adverse eīects.

Chapter 6 | Renewable energy outlook

197

Recent developments Renewables are steadily becoming a greater part of the global energy mix, in parƟcular in the power sector and in regions that have put in place measures to promote their deployment. Double-digit growth rates have been observed in the last decade for some renewable energy technologies and renewables are projected to conƟnue to grow strongly over the Outlook period to 2035, provided that the necessary support measures are kept in place. However, the situaƟon is nuanced across the three main energy uses: electricity, heat and transport. Electricity generaƟon from renewable sources is growing rapidly for most technologies͖ while renewable energy use for heat is growing more slowly and remains under-exploited. AŌer a period of rapid expansion, the rate of growth of biofuels use has recently slowed, due largely to adverse weather condiƟons that reduced harvests and increased feedstock prices, as well as sustainability concerns. Investment in renewable power generaƟon has also been rising steadily but it fell, for the Įrst Ɵme, in 2012. In part, this reŇects falling unit costs͖ but it is perhaps also a sign that the prospects for renewables are becoming more complex.

© OECD/IEA, 2013

In Europe, rapid expansion of renewable power generaƟon, parƟcularly wind and solar, has occurred in recent years, driven by the requirements of the European Union’s Renewable Energy DirecƟve and naƟonal targets. However, low rates of power demand growth and a diĸcult economic situaƟon raise doubts about the Ɵmelines of future investments and policymakers in several countries have started to express concerns about the aīordability of high shares of certain types of renewable power generaƟon. These concerns relate, parƟcularly, to higher than anƟcipated rates of deployment of solar photovoltaic (PV) systems, driven, in some countries, by generous and unlimited subsidy schemes and rapidly falling PV system cost. For example, Spain acted in 2010 to adjust over-generous renewables subsidies and, more recently, a moratorium has been put on further subsidies to renewables. DiĸculƟes about integraƟng high levels of variable renewables into the electricity system are also emerging in some European countries. In the United States, the market for renewables has been growing strongly, in large part due to the conƟnuaƟon of sƟmulus policies directed at renewable energy, such as the provision of cash grants (instead of a tax credit) of up to 30% of investment costs for eligible renewable energy projects (US Treasury 1603 Program). This programme expired at the end of 2012, but many projects were able to pre-qualify and will receive this support if completed by the end of 2016. An investment tax credit and producƟon tax credits also provided support for renewables in the United States, despite uncertainty over the future of the programmes. Indeed, doubts about their renewal at the end of 2012 led to high growth in that year͖ as developers pressed to complete projects in Ɵme to receive support. (Uncertainty surrounding future policy support measures has oŌen caused ͞boom and bust͟ cycles for capacity addiƟons of renewables). Renewable porƞolio standards, currently in eīect in 30 states and the District of Columbia, conƟnue to provide an important incenƟve to boost deployment. Along with blending mandates, annually increasing volume requirements under the Renewable Fuels Standard (RFS) have been a major driver for higher consumpƟon of biofuels each year since its enactment in 2005. 198

World Energy Outlook 2013 | Global Energy Trends

With rapidly growing power demand and concerns over energy security and local polluƟon, deployment of renewables has been acceleraƟng and is expected to conƟnue to do so in non-OECD countries. In China, the energy development plan, published in January 2013 as part of the 12th Five-zear Plan, sets ambiƟous renewables targets with mandatory 2015 targets for non-fossil energy use, energy intensity, carbon intensity and parƟculate emissions. India’s 12th Five-zear Plan foresees an increase in grid-connected renewable generaƟon capacity of 11ථGW from large hydropower and 30ථGW from other renewable sources by 2017. Major increases in renewables capacity are planned in the coming years in Brazil, led by hydropower, bioenergy and onshore wind (see Chapter 10). Tendering schemes in South Africa, the United Arab Emirates and Morocco are prompƟng investment in wind, solar PV and concentraƟng solar power (CSP), and many other countries with rising power demand are also embarking on large-scale deployment (IEA, 2013a).

1

AŌer global biofuels producƟon more than doubled between 2006 and 2010, driven by supporƟve policies in Brazil, the United States and the European Union, growth in 2011 and 2012 stagnated, despite high oil prices. A combinaƟon of physical and policy-related issues was to blame. Ethanol output in Brazil and the United States was aīected by poor sugarcane and corn harvests, leading to a lack of feedstock supply and high prices. In Europe, high feedstock prices and poor margins, as well as strong compeƟƟon from nonEuropean producers, posed challenges for biodiesel producers. Provision for the blending of more than 10% ethanol in the gasoline pool in the United States has raised technical and economic challenges, while doubts about the sustainability of biofuels producƟon in the European Union have led to a proposal to limit the use of food-crop derived biofuels to 6% of transport fuel. The producƟon of advanced biofuels – which oīer the prospect of requiring less land, improving greenhouse-gas balances and lower compeƟƟon between food and fuel – has been expanding, but only slowly.

6

The porƟon of modern renewable energy for heat in total Įnal heat demand has risen only slowly and is currently just above 10%. Most of this contribuƟon comes from bioenergy, although solar thermal and geothermal are playing an increasing part as they become progressively more cost compeƟƟve in a number of markets and circumstances. However, these technologies face disƟnct market and insƟtuƟonal challenges to deployment, with renewable heat receiving much less policy aƩenƟon than electricity from renewables or biofuels. To date, only 35 countries have policy frameworks supporƟve of renewable heat (mostly within the European Union stemming from the Renewables DirecƟve).

3 4 5

7 8 9 10 11 12 13 14 15

Renewables outlook by scenario

© OECD/IEA, 2013

2

There is a rapid increase in the use of renewable energy in each of the three scenarios presented in this Outlook (Tableථ6.1). This is primarily the result of the creaƟon of an environment, through policy, in which costs can be driven down so that renewable energy technologies become more compeƟƟve with other energy sources. In a limited, but growing, number of cases, they become fully compeƟƟve.

16

ReŇecƟng diīerences in the assumed level of policy acƟon across the scenarios, the share of renewables in total primary energy demand in 2035 varies markedly, from 26% in the

18

Chapter 6 | Renewable energy outlook

199

17

450ථScenario, to 18% in the New Policies Scenario and 15% in the Current Policies Scenario. By comparison, renewables met 13% of the world’s primary energy demand in 2011. Because renewables include both tradiƟonal and modern forms, its growth is the net result of two opposing trends. Dominant is a dramaƟc rise in demand for modern renewable energy (albeit from fairly low levels). The other is a shiŌ away from the use of tradiƟonal biomass – mostly fuel wood, charcoal, animal dung and agricultural residues used for heaƟng and cooking – in favour of modern forms, such as gas, electricity and liqueĮed petroleum gas (LPG). Reducing tradiƟonal biomass brings important health beneĮts by limiƟng exposure to local air pollutants (see Chapterථ2). In the New Policies Scenario, the share of tradiƟonal biomass in total primary energy demand drops from 5.7% in 2011 to 3.9% in 2035, as the reducƟon of tradiƟonal biomass use in Asia more than oīsets the populaƟon-driven increase in Africa. Table 6.1 ‫ ٲ‬World renewable energy use by type and scenario New Policies

© OECD/IEA, 2013

Primary energy demand (Mtoe) United States Europe China Brazil Share of global TPED Electricity generaƟon (TWh) Bioenergy Hydro Wind Geothermal Solar PV ConcentraƟng solar power Marine ^ŚĂƌĞŽĨƚŽƚĂůŐĞŶĞƌĂƟŽŶ Heat demand*(Mtoe) Industry Buildings* and agriculture ^ŚĂƌĞŽĨƚŽƚĂůĮŶĂůĚĞŵĂŶĚ Biofuels (mboeͬd)** Road transport AviaƟon*** Share of total transport TradiƟonal biomass (Mtoe) Share of total bioenergy ^ŚĂƌĞŽĨƌĞŶĞǁĂďůĞĞŶĞƌŐLJĚĞŵĂŶĚ

Current Policies

450 Scenario

2011

2020

2035

2020

2035

2020

2035

1 727 140 183 298 116 13% 4 482 424 3 490 434 69 61 2 1 20% 343 209 135 8% 1.3 1.3 2% 744 57% 43%

2 193 196 259 392 148 15% 7 196 762 4 555 1 326 128 379 43 3 26% 438 253 184 10% 2.1 2.1 4% 730 49% 33%

3 059 331 362 509 207 18% 11 612 1 477 5 827 2 774 299 951 245 39 31% 602 316 286 12% 4.1 4.1 0.1 6% 680 37% 22%

2 130 191 250 373 146 14% 6 844 734 4 412 1 195 114 352 35 3 24% 432 255 177 9% 1.9 1.9 3% 732 50% 34%

2 729 282 326 445 204 15% 10 022 1 250 5 478 2 251 217 680 122 24 25% 551 308 243 11% 3.3 3.2 0.1 4% 689 40% 25%

2 265 215 270 405 150 16% 7 528 797 4 667 1 441 142 422 56 3 28% 446 248 198 10% 2.6 2.6 5% 718 47% 32%

3 918 508 452 690 225 26% 15 483 2 056 6 394 4 337 436 1 389 806 64 48% 704 328 376 16% 7.7 6.8 0.9 15% 647 29% 17%

* Excludes tradiƟonal biomass. **ථExpressed in energy-equivalent volumes of gasoline and diesel. *** Includes internaƟonal bunkers. Note: Mtoe с million tonnes of oil equivalent͖ TPED с total primary energy demand͖ TWh с terawaƩ-hour͖ mboeͬd с million barrels of oil equivalent per day.

200

World Energy Outlook 2013 | Global Energy Trends

Renewables outlook by use in the New Policies Scenario

1

Renewables contribute an increasing share to total primary energy in the New Policies Scenario and reach 18% in 2035, with the share increasing for all uses and in almost all regions (Figureථ6.1). Demand for modern renewable energy – including hydropower, wind, solar, geothermal, marine and bioenergy – rises almost two-and-a-half Ɵmes, from 983ථmillion tonnes of oil equivalent (Mtoe) in 2011 to almost 2ථ400ථMtoe in 2035. Its share of total primary energy demand increases from 8% to 14%. A rapid uptake of hydropower occurs mainly in non-OECD countries, where signiĮcant resources remain untapped and oīer a cost-eīecƟve means of meeƟng fast-growing electricity demand. For most other technologies, the growth is driven by conƟnued support, although other factors, such as falling technology costs and, in some regions, rising fossil fuel prices and carbon pricing also contribute. TradiƟonal biomass remains an important energy source in parts of the world that conƟnue to lack access to clean cooking faciliƟes, although at the global level its use drops from 744ථMtoe in 2011 to 680ථMtoe in 2035. Figure 6.1 ‫ٲ‬

Renewable energy share in total primary energy demand by category and region in the New Policies Scenario, 2011 and 2035

50%

Addional market share in 2035

40%

2011

3 4 5 6 7 8 9

30%

10

20%

11

10%

12 World US EU China Electricity generaon

World US EU China Heat producon*

13

World US EU China Road transport

* Excludes tradiƟonal biomass. Note: US с United States͖ EU с European Union.

14

WŽǁĞƌŐĞŶĞƌĂƟŽŶ

© OECD/IEA, 2013

2

In the New Policies Scenario, renewables power generaƟon expands by over 7ථ000ථterawaƩhours (TWh) between 2011 and 2035. This is equivalent to around one-third of current global power generaƟon, and almost half of the projected increase in total power generaƟon to 2035 (see Chapter 5). The share of renewables in the global power mix rises from 20% in 2011 to 31% in 2035 (Tableථ6.2). CollecƟvely, renewables become the world’s second-largest source of power generaƟon before 2015 and approach coal as the primary source by the end of the period. There is rapid expansion of wind and solar PV, coupled with steady increases in both hydropower and bioenergy.

Chapter 6 | Renewable energy outlook

201

15 16 17 18

Table 6.2 ‫ ٲ‬Renewables-based electricity generation by region in the New Policies Scenario (TWh) Share of total generaƟon

Renewables generaƟon

Share of variable renewables* in total generaƟon

2011

2020

2030

2035

2011

2035

2011

2035

OECD

2 116

2 994

3 943

4 434

19.6%

33.8%

3.6%

14.2%

Americas

1 014

1 313

1 733

1 965

19.0%

29.6%

2.6%

11.0%

544

740

1 039

1 211

12.6%

23.0%

2.9%

10.7%

900

1 353

1 710

1 889

24.9%

45.2%

6.3%

21.0%

United States Europe Asia Oceania Japan Non-OECD E. EuropeͬEurasia Russia Asia

203

329

500

581

10.9%

25.5%

1.1%

10.9%

133

213

304

343

12.7%

28.2%

0.9%

11.4%

2 365

4 202

6 099

7 178

20.9%

29.9%

1.0%

7.8%

290

357

457

528

16.9%

21.8%

0.2%

2.3%

169

200

265

312

16.1%

20.5%

0.0%

1.1%

1 173

2 569

3 787

4 423

16.9%

27.2%

1.4%

9.1%

China

814

1 888

2 515

2 804

17.1%

28.0%

1.5%

9.9%

India

183

350

666

850

17.4%

25.2%

2.3%

10.4%

21

48

141

226

2.4%

12.9%

0.0%

6.8%

116

205

403

550

16.8%

36.0%

0.4%

5.6%

765

1 023

1 312

1 451

69.0%

71.0%

0.4%

6.2%

463

614

782

862

87.1%

79.5%

0.5%

8.9%

4 482

7 196

10 042

11 612

20.3%

31.3%

2.2%

10.0%

696

1 113

1 427

1 580

21.4%

43.8%

6.9%

23.1%

Middle East Africa LaƟn America Brazil World European Union

© OECD/IEA, 2013

* Variable renewables include solar PV and wind power.

Two-thirds of the increase in power generaƟon from renewables occurs in non-OECD regions, with these countries accounƟng for 62% of total renewables generaƟon in 2035, up from 53% in 2011. China alone accounts for 28%, or 1ථ990ථTWh, of the total growth in generaƟon from renewables, more than the European Union, United States and Japan combined (Figureථ6.2). Considerable growth is also seen in LaƟn America, India, Africa and Southeast Asia, mainly driven by policy intervenƟons. The increase in the United States, which contributes over 70% of the increase in its total generaƟon over the period, comes despite strong compeƟƟon from natural gas and also thanks to the decline in coal-Įred generaƟon. It is driven by federal tax credits and state-level renewable energy standards, which are assumed to conƟnue also beyond 2020. In the European Union, the increase in generaƟon from renewables far exceeds the increase in total generaƟon, as output falls from coal-Įred and nuclear plants. In Japan, mainly in response to the generous support policies recently put in place, electricity generaƟon from renewables increases by 160%, its share increasing from 13% in 2011 to 28% in 2035. Policy acƟon is also the main driver of growth in India, where ambiƟous targets have been set to scale-up renewable energy capacity in order to overcome electricity shortages and increase access. There is an eleven-fold increase in generaƟon from renewables in the Middle East, reŇecƟng the

202

World Energy Outlook 2013 | Global Energy Trends

region’s considerable solar and wind resources and growing recogniƟon of their potenƟal importance in saƟsfying rapid growth in power demand, while freeing up oil and natural gas for export. Figure 6.2 ‫ ٲ‬Incremental electricity generation from renewables in selected

TWh

regions, 2011-2035 ASEAN

Other renewables

Africa

Solar PV

1 800 1 500

Lan America

900 600

Wind

China

6

India

Hydro

7

Incremental renewable generaon by selected region

China by type

8

European Union

300

4 5

United States

1 200

More than 3ථ100ථgigawaƩs (GW) of renewables capacity are added over the period, equivalent to almost three Ɵmes the present total installed capacity of the United States. AŌer taking account of the reƟrement of older installaƟons, this results in installed capacity of renewables increasing by a factor of around 2.5, from almost 1ථ600ථGW in 2012 to nearly 4ථ000ථGW by 2035. Annual capacity addiƟons rise steadily over the period, with a brief downturn around 2020, when rapid expansion of hydropower in China slows as higherquality sites become scarcer (Figureථ6.3). Including the replacement for reƟring capacity, annual gross capacity addiƟons are around 180ථGW by the end of the projecƟon period, compared with over 115ථGW in 2012. Figure 6.3 ‫ ٲ‬Average annual increases in renewables-based capacity* by region in the New Policies Scenario GW

2 3

2 100

Japan

1

120

Brazil European Union

80

United States India

60

10 11 12 13 14

Rest of world

100

9

15 16

China

40

17

© OECD/IEA, 2013

20 1991 -1995

1996 -2000

2001 -2005

2006 -2012

2013 -2020

2021 -2025

2026 -2030

18

2031 -2035

* Excludes capacity that directly replaces reƟred capacity of the same technology type.

Chapter 6 | Renewable energy outlook

203

Renewable energy technologies make up more than 50% of gross capacity addiƟons in the power sector, pushing their share of installed power capacity from 28% in 2012 to 40% in 2035. Wind, with gross capacity addiƟons of almost 1ථ250ථGW, makes the largest contribuƟon to the growth, followed by solar PV (750ථGW) and hydro (740ථGW) (Figureථ6.4). Figure 6.4 ‫ ٲ‬Cumulative global renewables-based capacity additions

GW

and retirements by technology in the New Policies Scenario, 2013-2035 1 500

Addions

1 250

Rerements

1 000

Net addions

750 500 250 0 -250 -500

Wind

Hydro

Solar PV

Bioenergy

Other

Electricity generaƟon from bioenergy more than triples over the projecƟon period, with China, United States and the European Union accounƟng for over half of the growth. Its share of total generaƟon doubles from 2% to 4%. The share of hydropower in total power generaƟon stays stable throughout the Outlook period, at about 16%. Hydropower remains the leading source of renewables-based power, although its share of renewable electricity generaƟon falls from almost 80% today to around half in 2035, as the scope for further addiƟons is gradually reduced and other renewable technologies are deployed at a faster rate. Hydropower output rises from almost 3ථ500ථTWh in 2011 to 5ථ800ථTWh in 2035, based on an increase in installed capacity from 1ථ060ථGW to 1ථ730ථGW over the same period. The expansion is concentrated in non-OECD countries. China accounts for almost 25% of the increase in generaƟon, its capacity rising from 246ථGW in 2012 to 430ථGW in 2035. China added 16ථGW of new hydropower capacity in 2012 and further strong growth is projected unƟl around 2020, when growth slows as China gets closer to uƟlising its full potenƟal. Brazil also conƟnues to rely heavily on hydropower to meet electricity demand, adding around 70ථGW of capacity over the projecƟon period, to reach an installed capacity of 151 GW in 2035 (see Chapter 10). In the OECD, generaƟon from hydropower increases by a modest 16%, with the growth focused in North America and the European Union.

© OECD/IEA, 2013

Biofuels ConsumpƟon of biofuels is projected to rise from 1.3ථmillion barrels of oil equivalent per day (mboeͬd) in 2011 to 2.1ථmboeͬd in 2020, and 4.1ථmboeͬd in 2035 (Tableථ6.3). By 2035, biofuels meet 8% of total road-transport fuel demand, up from 3% today. Ethanol remains 204

World Energy Outlook 2013 | Global Energy Trends

the dominant biofuel, making up about three-quarters of global biofuels use throughout the period. ConsumpƟon of biodiesel in road transport more than triples over the Outlook period, to 1.1ථmboeͬd in 2035. Combined, the United States, Brazil, the European Union, China and India account for about 90% of world biofuels demand throughout the Outlook period, with government policies driving the expansion in these regions. These projecƟons are similar to those made in WEO-2012, despite a drop in investment in the sector last year and a temporary slowdown in producƟon growth, due primarily to poor harvests in the United States and Brazil. ConƟnued policy support and a return to normal harvests put biofuels consumpƟon back on track over the long term. In addiƟon to the use of biofuels in road transport, its use in aviaƟon begins to make inroads over the projecƟon period. Table 6.3 ‫ ٲ‬Ethanol and biodiesel consumption in road transport by region in

Biofuels total

Share of road transport energy use

2011

2035

2011

2035

2011

2035

2011

2035

OECD

0.7

1.5

0.2

0.8

0.9

2.3

4%

12%

Americas

0.6

1.3

0.1

0.3

0.7

1.6

4%

13%

United States

0.6

1.2

0.1

0.3

0.7

1.5

5%

15%

Europe

0.0

0.2

0.2

0.5

0.2

0.7

4%

12%

Non-OECD

0.3

1.4

0.1

0.4

0.4

1.8

2%

5%

E. EuropeͬEurasia

0.0

0.0

0.0

0.0

0.0

0.0

0%

2%

Asia

0.0

0.7

0.0

0.1

0.1

0.8

1%

4%

China

0.0

0.4

0.0

0.0

0.0

0.4

1%

4%

India

0.0

0.2

0.0

0.0

0.0

0.2

0%

4%

0.3

0.8

0.1

0.2

0.4

1.0

10%

20%

0.2

0.6

0.0

0.2

0.3

0.8

19%

30%

World

1.0

2.9

0.4

1.1

1.3

4.1

3%

8%

European Union

0.0

0.2

0.2

0.5

0.2

0.7

5%

15%

LaƟn America Brazil

© OECD/IEA, 2013

Biodiesel

The United States remains the largest biofuels market, spurred on by the Renewable Fuel Standard (RFS) through 2022 and assumed conƟnuaƟon of support thereaŌer, with consumpƟon increasing from around 0.7ථmboeͬd to 1.5ථmboeͬd in 2035, by which Ɵme biofuels meet 15% of road-transport energy needs. Driven by blending mandates and strong compeƟƟon between ethanol and gasoline, Brazil remains the second-largest market and conƟnues to have a larger share of biofuels in its transport fuel consumpƟon than any other country. In 2035, biofuels meet 30% of Brazilian road-transport fuel demand up from 19% today. Supported by the Renewable Energy DirecƟve and conƟnued policy support, biofuels use in the European Union more than triples over the period to 0.7ථmboeͬd in 2035, represenƟng 15% of road-transport energy consumpƟon. In China, government plans for expansion lead to demand for biofuels reaching 0.4ථmboeͬd in 2035, many Ɵmes the Chapter 6 | Renewable energy outlook

2 3 4 5 6

the New Policies Scenario (mboe/d) Ethanol

1

205

7 8 9 10 11 12 13 14 15 16 17 18

current level. India established an ambiƟous NaƟonal Mission policy on biofuels in 2009, but the infancy of the ethanol industry and diĸculty in meeƟng current targets constrains future demand growth in the projecƟons.

© OECD/IEA, 2013

The outlook for biofuels is highly sensiƟve to possible changes in government subsidies and blending mandates, which remain the main sƟmulus for biofuels use. Over the past year much uncertainty has developed about how biofuel policies in several key markets will evolve. At the Ɵme of wriƟng, discussions in the European Union were conƟnuing on the possible introducƟon of a 6% cap on the amount of convenƟonal biofuels that can be counted towards the level of renewable energy in transport mandated in the Renewable Energy DirecƟve. These discussions are driven by sustainability issues, including concern that feedstock producƟon for biofuels contributes to deforestaƟon or pre-empts land that could be used to grow food. The European Union has also placed temporary anƟ-dumping duƟes on biofuel imports from the United States, ArgenƟna and Indonesia, with material impact on trade in biofuels, so clouding the picture for future trade. In the United States, a review of the federal RFS is underway, which could signiĮcantly alter the long-term outlook for ethanol, amid widespread concerns that the supply targets for 2022 are not achievable. One key concern is the amount of ethanol that can be consumed by vehicles on the road (oŌen referred to as the ͞blend wall͟), due to strong resistance from various parts of the industry to blending levels higher than 10% (E10) and logisƟcal barriers to supplying the current Ňex-fuel vehicle Ňeet with high-ethanol content fuels, such as E85. A second concern is whether domesƟc producƟon of cellulosic biofuels can meet oĸcial volume goals, as cellulosic biofuel supply targets have had to be lowered in the past few years. On the other hand, Brazil has made policy changes over the last year poinƟng to higher growth for biofuels, including restoring the ethanol blending mandate to 25%, aŌer reducing it to 20% in late 2011 due to poor sugarcane harvests. Advanced biofuels oīer the prospect of increasing biofuels supply while reducing or eliminaƟng sustainability concerns for biofuels. Cellulosic ethanol is a promising advanced biofuel that can be derived from a variety of feedstocks, including bagasse and agricultural residues, as well as dedicated energy crops. Much work on advanced biodiesel at present is concentrated on the use of feedstocks with far higher yields than convenƟonal feedstock, including palm oil, rapeseed and jatropha. But all the feedstocks depend on conversion technologies that are mainly in the research and development, pilot or demonstraƟon phases. If developed successfully, they hold the promise of achieving lower overall unit costs and imposing lower land requirements than convenƟonal biofuels. While a few commercial-scale units and about 100 plants at pilot or demonstraƟon scale already exist, widespread deployment will require lower costs which further technological progress could bring (IEA 2013b). Because of the lack of commercial scale producƟon of advanced biofuels, the supply mandate for cellulosic biofuels under the RFS in the United States was reduced again in 2013. In the New Policies Scenario, advanced biofuels become available at commercial scale around 2020, with their share of total biofuels supply rising from below 1% today to almost 20% in 2035, led by the United States, Europe, China and Brazil. 206

World Energy Outlook 2013 | Global Energy Trends

Heat Heat is the largest energy service demand worldwide, typically used for process applicaƟons in industry, and for space and water heaƟng, and cooking in the buildings sector. The energy use required to meet this service demand accounts for around half of total Įnal energy consumpƟon. Currently, most of the contribuƟon of renewables to heat producƟon comes from biomass used in tradiƟonal ways for cooking and heaƟng in developing countries (Figureථ6.5). The use of tradiƟonal biomass for heat amounted to 744ථMtoe in 2011 and made up 18% of total global energy use for heat. Such use is oŌen unsustainable because of the low eĸciency with which the fuels are converted, the emissions produced (leading to potenƟal health problems) and the diĸculty in maintaining supply. More modern and eĸcient technologies uƟlising renewable energy – (non-tradiƟonal) bioenergy, geothermal and solar thermal in parƟcular – are playing an increasing role in heat supply and met 8% of total global demand for heat in 2011. Figure 6.5 ‫ ٲ‬Share of renewables in heat production in the residential sector for selected regions in the New Policy Scenario 20%

40%

60%

80%

100%

Africa 2011 2035

2 3 4 5 6 7 8 9

India 2011 2035

10

China 2011 2035 European Union 2011 2035

11

United States 2011 2035

12

OECD Asia 2011 Oceania 2035

13 Space heang

© OECD/IEA, 2013

1

Water heang

Cooking

Tradional biomass

In the residential sector, more than 40% of the heat supplied globally today by modern renewables is consumed in Europe, mainly in the form of bioenergy for space heating. The United States and China account for 14% and 11% of modern renewable use for heat respectively. Recent growth in China has outpaced all other regions. In the last five years, China accounted for almost 40% of global growth in the use of modern renewable energy for heat in the residential sector, driven by the rapid deployment of solar water heaters, which are increasingly cost competitive with conventional fuels (Eisentraut and Brown, 2013), and household biogas systems. In industry, 70% of renewable energy use for heat is in the light industry sector such as food, tobacco and machinery. Almost all of it is bioenergy, which accounts for 11% (136ථMtoe) of global light industry’s total energy demand. Chapter 6 | Renewable energy outlook

207

14 15 16 17 18

Global modern renewable energy use for heat producƟon increases by 75% in the New Policies Scenario, reaching 600ථMtoe in 2035. By the end of the period, modern renewables meet 12% of total heat demand, compared with 8% in 2011. The use of tradiƟonal biomass for heat producƟon falls some 10%, to 680 Mtoe in 2035. It conƟnues to be the main source of heat in the residenƟal sector in many developing countries, although in others the switch to modern energy services is made possible by rising incomes, ongoing urbanisaƟon and programmes to foster access to modern energy sources. Demand for modern renewables for heat in the residenƟal sector almost doubles, growing from 88ථMtoe in 2011 to 165ථMtoe by 2035. Most of the growth occurs in China and Europe, with modern bioenergy remaining the dominant source even though solar and geothermal both grow at much faster rates, largely driven by the use of solar thermal heaters in China.

Focus on power generaƟon from variable renewables Unlike dispatchable power generaƟon technologies, which may be ramped up or down to match demand, the output from solar PV and wind power is Ɵed to the availability of the resource.1 Since their availability varies over Ɵme, they are oŌen referred to as variable renewables, to disƟnguish them from the dispatchable power plants (fossil fuel-Įred, hydropower with reservoir storage, geothermal and bioenergy). Wind and solar PV power are not the only variable renewables – others include run-of-river hydropower (without reservoir storage) and concentraƟng solar power (without storage) – but PV and wind power are the focus of this secƟon as they have experienced parƟcularly strong growth in recent years and this is expected to conƟnue. The characterisƟcs of variable renewables have direct implicaƟons for their integraƟon into power systems (IEA, forthcoming 2014a). The relevant properƟes include: „ Variability: power generaƟon from wind and solar is bound to the variaƟons of the

wind speed and levels of solar irradiance. „ Resource locaƟon: good wind and solar resources may be located far from load

centres. This is parƟcularly true for wind power, both onshore and oīshore, but less so for solar PV, as the resource is more evenly distributed. „ Modularity: wind turbines and solar PV systems have capaciƟes that are typically on

the order of tens of kilowaƩs (kW) to megawaƩs (MW), much smaller than convenƟonal power plants that have capaciƟes on the order of hundreds of MW. „ Uncertainty: the accuracy of forecasƟng wind speeds and solar irradiance levels

© OECD/IEA, 2013

diminishes the earlier the predicƟon is made for a parƟcular period, though forecasƟng capabiliƟes for the relevant Ɵme-frames for power system operaƟon (i.e. next hours to day-ahead) are improving.

1.ഩElectricity generation from (non-dispatchable) variable renewables, such as wind and solar, is weather dependent and can only be adjusted to demand within the limits of the resource availability. 208

World Energy Outlook 2013 | Global Energy Trends

„ Low operaƟng costs: once installed, wind and solar power systems generate electricity

at very low operaƟng costs, as no fuel costs are incurred. „ Non-synchronous generaƟon: power systems are run at one synchronous frequency:

most generators turn at exactly the same rate (commonly 50 Hz or 60ථHz), synchronized through the power grid. Wind and solar generators are mostly non-synchronous, that is, not operaƟng at the frequency of the system. The extent to which these properƟes of variable renewables pose challenges for system integraƟon largely depends on site-speciĮc factors, such as the correlaƟon between the availability of wind and solar generaƟon with power demand͖ the Ňexibility of the other units in the system͖ available storage and interconnecƟon capacity and the share of variable renewables in the overall generaƟon mix. The speed at which renewables capacity is introduced is also important, as this inŇuences the ability of the system to adapt through the normal investment cycle. EīecƟve policy and regulatory design for variable renewables needs to co-ordinate the rollout of their capacity with the availability of Ňexible dispatchable capacity, grid maintenance and upgrades, storage infrastructure, eĸcient market operaƟon design, as well as public and poliƟcal acceptance.

Wind power GeneraƟng power from wind turbines varies with the wind speed. Although there are seasonal paƩerns in some regions, the hourly and daily variaƟons in wind speed have a less predictable, stochasƟc paƩern. Geographically, good wind sites are typically located close to the sea, in Ňat open spaces andͬor on hills or ridgelines, but the suitability of a site also depends on the distance to load centres and site accessibility.

© OECD/IEA, 2013

For onshore wind turbines, capacity factors – the raƟo of the average output over a given Ɵme period to maximum output – typically range from 20% to 35% on an annual basis͖ excellent sites can reach 45% or above. The power output from new installaƟons is increasing, as turbines with larger rotor diameters and higher hub heights (the distance between the ground and the centre of the rotor) can take advantage of the increased wind speeds at higher alƟtudes. Moreover, wind projects are increasingly being tailored to the characterisƟcs of the site by varying the height, rotor diameter and blade type. Wind turbines that are able to operate at low wind speeds oīer the advantage of a steadier generaƟon proĮle, reducing the variability imposed upon the power system, but likely reducing annual generaƟon. Wind turbines located oīshore can take advantage of stronger and more consistent sea breezes. Wind speeds tend to increase with increasing distance from the shore, but so too does the seaŇoor depth, requiring more complex foundaƟon structures. Capacity factors are generally higher ranging from 30% to 45% or more, as distance from the shore or hub height increases. However, oīshore wind turbines are more expensive to install because of the high costs associated with the foundaƟons and oīshore grid connecƟons. BoƩlenecks can also occur due to a shortage of specialised installaƟon vessels. Chapter 6 | Renewable energy outlook

209

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18

ZĞĐĞŶƚƚƌĞŶĚƐĂŶĚƉƌŽũĞĐƟŽŶƐ AŌer experiencing growth of around 25% per year over the past decade, wind power made up 2.3% of global power generaƟon in 2012. Globally, wind capacity rose by 44ථGW in 2012, a record year, to 282ථGW. This is almost Įve Ɵmes the capacity in place in 2005. New installaƟons were concentrated in China (adding 13ථGW in 2012) and the United States and the European Union (both adding 12ථGW). There was a surge in installaƟons in the United States, as developers sought to secure producƟon tax credits, which were set to expire at the end of 2012. AddiƟons in China were 5ථGW lower than in 2011 due to boƩlenecks for connecƟons to the grid. Out of this total, oīshore wind saw a 32% increase in global installed capacity in 2012, a rise of 1.3ථGW to 5.4ථGW, the highest annual capacity addiƟon to date. Some 90% of this was added in the European Union, mainly in the United
Scenario 400

40%

300

30%

Wind offshore Wind onshore

2012 2020 2035

2012 2020 2035

10%

2012 2020 2035

100

2012 2020 2035

20%

2012 2020 2035

200

United States

European Union

China

India

Rest of world

Share of global wind capacity (right axis)

© OECD/IEA, 2013

Solar photovoltaics Power generaƟon from solar PV installaƟons varies with the level of solar irradiaƟon they receive. Geographically, solar irradiaƟon increases with proximity to tropical regions and is more uniformly distributed than wind. Seasonal and daily paƩerns in output from solar PV systems can be fairly well forecast – on a clear day, solar generaƟon follows a bell shape, 210

World Energy Outlook 2013 | Global Energy Trends

reaching its maximum around midday – but there remains an element of unpredictability, such as the extent of cloud cover or interference through snow, sand or dust cover.

1

Capacity factors vary widely, but generally lie within 10% and 20%, or above. The last ten years brought important technology progress, with signiĮcant cost reducƟons. Newer technologies, such as thin Įlm technologies, are gaining growing market shares and bring further potenƟal for cost reducƟons. Systems which include sun tracking systems can reduce variability, as can an array of panels with diīering orientaƟons, but in both cases costs are increased.

2

ZĞĐĞŶƚƚƌĞŶĚƐĂŶĚƉƌŽũĞĐƟŽŶƐ

5

Solar PV generaƟon expanded by 50% per year worldwide over the last decade, reaching almost 100ථTWh in 2012. In this year, total installed capacity of solar PV increased by 43%, or 29.4ථGW, represenƟng 15% of the total growth in global power generaƟon capacity. Germany alone, under the impetus of strong government support, accounted for more than one-quarter of the increase with 7.6ථGW of addiƟons. Other countries with major addiƟons include Italy (3.6ථGW), China (3.5ථGW), United States (3.3ථGW), Japan (2.0ථGW) and India (1.1ථGW). In each country, the growth was driven by government support programmes and subsidies.

Scenario

6 7 8

10

60%

Share of global solar PV capacity (right axis)

100

40%

12

50

20%

13

2012 2020 2035

150

2012 2020 2035

Solar PV

2012 2020 2035

80%

2012 2020 2035

200

2012 2020 2035

GW

4

9

Figure 6.7 ‫ ٲ‬Installed solar PV capacity by region in the New Policies

© OECD/IEA, 2013

3

United States

European Union

China

India

Rest of world

14 15

In the New Policies Scenario, electricity produced from solar PV rises to 950ථTWh in 2035, as its share of global electricity generaƟon increases from 0.4% to 2.6%. This is underpinned by a seven-fold increase in installed solar PV capacity over the Outlook period, reaching 690ථGW in 2035 (Figureථ6.7). GeneraƟon from solar PV increases faster than installed capacity due to technical improvements and deployment in regions with high quality resources. Solar PV on buildings accounts for the majority of installaƟons, its share declining over the Outlook period as large-scale faciliƟes operated by uƟliƟes gain Chapter 6 | Renewable energy outlook

11

211

16 17 18

market share. Driven by big increases in China (150ථGW) and India (90ථGW), non-OECD regions account for almost 60% of the increase in global solar PV capacity. Large increases also occur in the European Union and the United States (both around 80ථGW), and Japan (50ථGW). Through ongoing reducƟons, generaƟon costs become comparable to retail electricity prices in several countries, but growth of solar PV will conƟnue to be closely linked to the provision of government subsidies as, over the course of the Outlook period, solar PV is expected to become compeƟƟve in only a limited number of circumstances when compared to the average wholesale electricity price (Spotlight).

/ŵƉůŝĐĂƟŽŶƐĨŽƌĞůĞĐƚƌŝĐŝƚLJƐLJƐƚĞŵƐĂŶĚŵĂƌŬĞƚƐ The impact of a growing component of variable renewables on the power system depends on the Ɵming and co-ordinaƟon of renewables capacity addiƟons, the investment cycles in the power system and the rate of deployment of measures to facilitate their integraƟon into the system. The main impacts of locaƟon constraints and modularity are on the transmission and distribuƟon network, while variability and uncertainty impact the way other power plants in the mix are operated.

/ŵƉůŝĐĂƟŽŶƐĨŽƌŐƌŝĚƐ The locaƟon of good variable renewable resources can be remote from demand centres, making transmission grid extensions necessary. Early and integrated planning of transmission corridors is necessary to maximise use of good resources and reduce public opposiƟon. In some locaƟons, transmission corridors will have to cross state or naƟonal borders, requiring co-operaƟon between transmission system operators and regulators.

© OECD/IEA, 2013

The transmission system costs involved to connect and integrate variable renewables depend on the distance to be covered, the status of development of the exisƟng grids and the amount of capacity of variable renewables to be integrated. Costs range between $100 and $250 per kW of added variable renewables capacity (Dena, 2010͖ EnerNex,ථ2011͖ NREL, 2010). In total, about $170ථbillion or some 10% of the global investment in transmission grids in the New Policies Scenario is required to extend the grid to accommodate the growth in renewables. The amount varies signiĮcantly by region. In Europe and Japan, high levels of deployment mean that the integraƟon of renewables accounts for a share of overall transmission investment of about 25% and 20% respecƟvely. The comparable Įgure is about 10% in the United States, China and India. The modularity of variable renewables can also have signiĮcant impacts on distribuƟon grid needs. Bypassing the high-voltage transmission grids that transport power from large convenƟonal power plants, wind and solar generators are typically connected at the distribuƟon level (wind at mid-voltage and solar mainly at low-voltage). At low levels of installed wind and solar capacity, their generaƟon can be consumed close to the producƟon site (especially for solar PV) and may reduce the strain on distribuƟon grids. At higher levels, the capacity of the distribuƟon grid may need to be raised to accommodate increasing volumes of electricity sold back to the grid by distributed generators. Voltage transformers can be an iniƟal boƩleneck͖ a need to upgrade line capaciƟes may follow. 212

World Energy Outlook 2013 | Global Energy Trends

The amount of investment to upgrade distribuƟon grids also depends on their current condiƟon. If these grids are in need of refurbishment, the addiƟonal costs may be low. In France and Germany, for example, each kilowaƩ of new variable renewables capacity will add an esƟmated $100 to $300 to the costs of the distribuƟon grids (LƂdl, et al., 2010; Dena, 2012; CRE, 2012). In the New Policies Scenario, total investment in distribuƟon grids to accommodate variable renewables amounts to over $90ථbillion globally, or about 2% of total distribuƟon investment. Bringing together transmission and distribuƟon (TΘD) costs aƩributable to variable renewables, the addiƟonal investment is about $260ථbillion, or some 4% of total TΘD investment over the Outlook period.

1

/ŵƉůŝĐĂƟŽŶƐĨŽƌĚŝƐƉĂƚĐŚĂďůĞƉŽǁĞƌƉůĂŶƚƐ

5

In the absence of a widespread uptake of the measures available to alleviate the challenges posed by variables renewables (Boxථ6.1), an increase in generaƟon from wind and solar power has implicaƟons for the operaƟon and use of dispatchable plants as well as for investment in such plants. Box 6.1 ‫ ٲ‬Reducing the challenges posed by variable renewables A number of operaƟonal and infrastructure measures can be taken to address the challenges posed by variable renewables. These include: „ AdapƟng the operaƟon of power systems. This can include the applicaƟon of

3 4

6 7 8 9

advanced forecasƟng techniques, and adapƟng the market and power plant dispatch rules, for example reducing the Ɵme between the commitment of power plants to generate electricity and real-Ɵme operaƟon.

10

„ Extending the transmission grid to capture remote resources and increase cross-

11

border trade, so as to reduce the eīects of variaƟons in solar irradiaƟon and wind speed on the system. This can be especially eīecƟve for wind (Schaber, et al., 2012). „ PromoƟng demand-side integraƟon. Modifying electricity demand according to

the variable supply could reduce the system impacts of wind and solar and also avoid the need for other integraƟon measures. „ InvesƟng in storage (such as pumped hydro storage, compressed air, hydrogen

or baƩeries). If deployed on a small scale (such as baƩeries for solar PV), storage can help to sustain reliance on local generaƟon and defer grid investment (IEA, 2014b). „ Balancing ŇuctuaƟons from variable renewable output with Ňexible forms of

generaƟon, such as gas turbines. „ Curtailing extreme wind and solar power generaƟon peaks, when variable

renewables output is very high compared to electricity demand, to reduce the ramping up and down of power output from other sources (Baritaud, 2012). © OECD/IEA, 2013

2

While all measures may be advantageous individually, co-ordinaƟon between the integraƟon measures is needed to maximise their beneĮts.

Chapter 6 | Renewable energy outlook

213

12 13 14 15 16 17 18

Electricity demand varies considerably during the course of a day, but it generally follows a predictable proĮle. For example, on a weekday demand may peak in the early evening as people arrive home and be lowest during the early hours of the morning when most people are asleep. However, generaƟon from wind and solar power is Ɵed to the availability of their resources and is oŌen not well matched with the electricity demand proĮle. The paƩern of the remaining electricity demand, aŌer variable renewables producƟon has been taken into account, also called residual electricity demand, can diīer markedly from the total electricity demand (Figureථ6.8). The variability of wind and solar generaƟon alters the peaks and troughs in the residual demand proĮle which requires the dispatchable plants to adjust their output level accordingly. However, where variable renewables generaƟon is well correlated with electricity demand (e.g. solar PV coinciding with air condiƟoning loads at midday) their generaƟon paƩern – up to a certain level of deployment – may be advantageous to the system by smoothing the demand proĮle. The greater the variability of residual demand, the greater the Ňexibility of dispatchable power plants must be to be able to respond to changes not only of demand but also to supply side changes. This can raise their operaƟonal costs (through not running at opƟmal eĸciency) and increase the wear-and-tear of power plant components. These ͞balancing costs͟ vary from system to system, depending on the presence of storage, the Ňexibility of the power plant Ňeet and also the quality of wind and solar resources and forecasts. Figure 6.8 ‫ ٲ‬Indicative hourly electricity demand and residual electricity demand with expanding deployment of solar PV (a) With PV at 25% of the peak demand

(b) With PV at 50% of the peak demand

(c) With PV at 100% of the peak demand

GW

60

Solar PV generaon Residual demand Electricity demand

50 40 30 20

© OECD/IEA, 2013

4PM

8PM

8AM

12PM

4AM

12AM

8PM

4PM

8AM

12PM

4AM

12AM

4PM

8PM

12PM

8AM

4AM

12AM

10

In regions where the electricity generaƟon of variable renewables increases faster than demand, uƟlisaƟon of exisƟng plants is reduced. In the New Policies Scenario, total wind capacity increases by around 850ථGW and solar PV by almost 600ථGW in the period to 2035 with about 40% of this increase occurring before 2020 (in the 450 Scenario wind and solar capacity increases by 1ථ400ථGW and some 900ථGW unƟl 2035, respecƟvely ΀Boxථ6.2΁). UnƟl 2020, many of the exisƟng dispatchable power plants will conƟnue to be needed, but will

214

World Energy Outlook 2013 | Global Energy Trends

likely experience less use, especially in regions that see major expansions of wind and solar generaƟon, such as Europe. In countries with fast-growing power demand, such as China, this eīect is less pronounced. Despite the increasing capacity of wind and solar, their variable and uncertain generation profile mean that the need for dispatchable capacity is not reduced significantly. The reason is that the share of installed wind and solar capacity that can be confidently relied upon at times of high demand is much lower than for dispatchable plants. This share is referred to as ͞capacity credit͟. It depends on the respective correlation of wind and solar supply with the load profile and the level of penetration of variable renewables. 2 For example, in the European Union, it typically falls between 5% and 10% for wind and 0% to 5% for solar PV. In the New Policies Scenario, wind and solar account for about 19% of global installed power capacity in 2035, reaching almost 35% in the European Union (Figureථ6.9). However, globally they contribute only about 2% to Įrm capacity (capacity that can be relied upon to generate electricity at any given Ɵme). The provision of suĸcient dispatchable capacity can entail addiƟonal costs. Assuming that addiƟonal gas turbines are used to meet this requirement, these adequacy costs are esƟmated to range from between $3-5 for each megawaƩ-hour (MWh) of addiƟonal generaƟon from variable renewables (IEA, 2011). Since the use of other power plants declines with increasing levels of variable renewables, the capacity mix gradually shiŌs to less capital-intensive power plant types, such as gasĮred power plants, for which proĮtability at low uƟlisaƟon rates is easier to achieve.

1 2 3 4 5 6 7 8 9 10

Figure 6.9 ‫ ٲ‬Shares of wind and solar power capacity and generation in the

11

New Policies scenario 35%

Share of variable renewables in total capacity

30% 25%

of which firm capacity*

20%

Share of variable renewables in total generaon

15%

12 13 14

10%

15

5% 2012 2020 2035 United States

2012 2020 2035 European Union

16

2012 2020 2035 China

© OECD/IEA, 2013

*ථFirm capacity of wind and solar is computed based on the capacity credit.

17

2.ഩ For solar, the capacity credit can be higher in systems where peaks in electricity demand are driven by demand for air conditioning, for example. Through interconnection over larger geographic areas, smoothing can be achieved, and the capacity credit can also be raised.

Chapter 6 | Renewable energy outlook

215

18

Box 6.2 ‫ ٲ‬Variable renewables in the 450 Scenario The stronger deployment of renewable energy technologies is one of the key features of the 450ථScenario. By 2035, their share of global power generaƟon increases to 48%, compared to 31% in the New Policies Scenario. The global share of wind and solar power generaƟon in the 450 Scenario increases to 18% in 2035 (compared to 10% in the New Policies Scenario), with important implicaƟons for the power system. Total wind and solar capacity reaches 2ථ700ථGW, which corresponds to 50% of peak demand in 2035. At a regional level, the capacity of variable renewables compared to peak load can be considerably higher; for example, in Europe it is more than 90% of peak demand and in China and Japan about 60% of peak demand. This means that the likelihood of momentary regional excess supply increases, when wind and solar generate electricity at their full capacity. In that case, there would be an important challenge for stable operaƟon of the system, due to the non-synchronous generaƟon of wind and solar power. A soluƟon, such as keeping online a share of thermal generators at all Ɵmes, would have to be considered. The commercial viability of these other power plants is a challenge, since they would operate at a very low uƟlisaƟon rate. Moreover, addiƟonal investments would be required in TΘD grids, as well as in other integraƟon measures.

/ŵƉůŝĐĂƟŽŶƐĨŽƌŵĂƌŬĞƚƉƌŝĐĞĨŽƌŵĂƟŽŶ In most liberalised electricity markets, spot wholesale prices are largely determined by the operaƟonal costs of the most expensive generaƟng unit used. Whenever low marginal cost power from wind and solar is fed into the system, generators with high operaƟng costs, at the upper end of the merit order,3 are needed less and the wholesale electricity price is, in consequence, lowered. Electricity end-users might beneĮt from this decrease depending on how much of the cost subsidies to renewables is passed through to them (see Chapterථ8). The merit order eīect may also reduce proĮt margins for all power generators, to the point that some generators become unproĮtable. This has been observed recently, for example, in some European markets, and has put in quesƟon whether some uƟliƟes will be able to recover the investment costs of dispatchable plants under current market condiƟons. This could potenƟally jeopardise the reliability of power supply if the situaƟon worsens.

© OECD/IEA, 2013

Market reforms have been introduced or are under consideraƟon in several countries where there is concern that price signals resulƟng from this eīect may not be suĸcient to sƟmulate Ɵmely and suĸcient investment in new dispatchable power plants or to maintain older plants in operaƟon. The opƟons include diīerent forms of capacity remuneraƟon or regulatory obligaƟon to maintain strategic reserve capacity or to allow hourly wholesale prices to increase unconstrained during Ɵmes of scarcity (for example, when peak demand periods coincide with limited generaƟon from variable renewables). Discussion of these issues remains open. One possibility is to incorporate measures which can reduce capacity needs, such as storage or demand-side management. 3.ഩ The merit order ranks the different generating units that are available in a power market in terms of their marginal cost of generation. It is often used to determine which units will be used to supply expected demand, with the cheapest units being used first. 216

World Energy Outlook 2013 | Global Energy Trends

ŽŵƉĞƟƟǀĞŶĞƐƐĂŶĚƵŶŝƚĐŽƐƚƐ The cost of producing electricity from solar PV and wind has fallen dramaƟcally over the last decade, leading to debate about whether they are now compeƟƟve, without subsidies, with the costs of power generated from fossil fuels. When measuring the compeƟƟveness of diīerent renewable energy technologies, it is important to disƟnguish between generaƟng power for sale and power produced by households for auto-consumpƟon.4 The laƩer typically involves solar PV.

ŽŵƉĞƟƟǀĞŶĞƐƐŽĨǀĂƌŝĂďůĞƌĞŶĞǁĂďůĞƐĂƐĂǁŚŽůĞƐĂůĞƉŽǁĞƌŐĞŶĞƌĂƟŽŶƐŽƵƌĐĞ For electricity produced to sell on wholesale markets, it is usually considered that breakeven is achieved when the levelised cost of electricity (LCOE)5 of a technology does not exceed the average wholesale electricity price received for generaƟon over its lifeƟme. Variable renewables, however, have limited or no means to adjust their power output across the day to maximise their revenues. At increasing rates of penetraƟon in the power mix, the price that they would receive in the market is likely to decrease over Ɵme, due to the socalled merit order eīect (Hirth, 2013; Mills and Wyser, 2012). One consequence of decreasing prices over Ɵme is that the support needed would be higher than currently calculated in the New Policies Scenario, where the benchmark is the average annual wholesale price. If the reference price considered for the calculaƟons of the wind and solar PV subsidies were to be lower than the annual average by 10%, the subsidies for wind and solar PV would be 12% higher. With a price 20% lower, they would be 25% higher. For the end-user, the higher cost of subsidies would, at least in part, be compensated by the lower wholesale prices.

© OECD/IEA, 2013

ŽŵƉĞƟƟǀĞŶĞƐƐŽĨƐŽůĂƌWsĨŽƌŚŽƵƐĞŚŽůĚƐ

1 2 3 4 5 6 7 8 9 10 11

For household auto-consumers, the break-even point for solar PV has typically been considered to be when the cost to the consumer reaches ͞grid parity͟ or ͞socket parity͟, that is, the point at which the levelised cost of electricity (excluding subsidies) falls to the average retail price for electricity. However, this approach has shortcomings and we quesƟon whether it is the appropriate metric to evaluate the compeƟƟveness of solar PV in households. An alternate approach, that takes account of other relevant costs, is to measure compeƟveness on the basis of ͞cost parity͟. This is the break-even point for the costs incurred, on the one hand, by a household with a solar PV system and, on the other hand, by a corresponding household that is solely reliant on the grid (Spotlight).

12

The disƟncƟon between grid parity and cost parity has important real-world implicaƟons. In most markets, the Įxed costs are only parƟally recovered through a Įxed component in the electricity bills and the remaining part (oŌen larger) is recovered through the

16

4.ഩ Auto-consumers are defined as those households which generate principally for their own consumption, with any excess being sold to the grid. 5.ഩ The levelised cost of electricity represents the average cost of producing electricity from a given technology, including all fixed and variable costs, expressed in terms of the present value equivalent.

Chapter 6 | Renewable energy outlook

217

13 14 15

17 18

variable component. From the single household perspecƟve, under such an electricity tariī structure, it might, therefore, be economically aƩracƟve to invest in PV, where grid parity is reached. This could lead to a signiĮcant addiƟonal amount of PV installaƟons. However, from a system perspecƟve, this creates a free-rider eīect, where households with PV systems do not pay fully for their share of the system’s Įxed costs, shiŌing the burden to households without PV systems. This could concentrate Įxed costs on fewer households, raising the retail prices against which the compeƟƟveness of PV systems is measured according to grid parity. These system level issues require thorough assessment and aƩenƟon from policymakers, regulators and retailers, who may need to consider the use of Ɵme-based metering and pricing, and tariīs adjusted to user proĮles to ensure both the full recovery and fair allocaƟon of system costs.

S P O T L I G H T Is residential solar PV already competitive? A fall of over 40% in the price of solar panels since 2010 has led some parƟes to make the case that electricity generated from residenƟal solar PV installaƟons has become – or is fast becoming – compeƟƟve with electricity generated from fossil fuels. These arguments have oŌen been based on the concept of ͞grid parity͟. But is grid parity the right criterion to measure the full compeƟƟveness of residenƟal PV, aŌer which it can stand on its own without the need for subsidies? The short answer is no, at least for households that remain connected to the grid. The reason is that Įxed system costs are not included in the calculaƟons of grid parity. The Įxed costs of a power system include costs such as the construcƟon and maintenance of the transmission and distribuƟon grids, metering and billing. From a system perspecƟve, these costs always need to be recovered. When allowance is made for these costs, the cost of generaƟon from solar PV systems would have to fall below grid parity to become compeƟƟve. Take an example, in which residenƟal solar PV has just reached grid parity. In the Įrst case, the household does not install solar PV. It pays $300 per year in Įxed charges (assuming all Įxed costs are passed through) and another $400 per year for the 4ථMWh it consumes, to give an average retail price of $175ͬMWh (Figureථ6.10a).

© OECD/IEA, 2013

In the second case, the household installs a solar PV system which produces 1.6ථMWh for consumpƟon on site, for a total cost of $280 (equal to 1.6ථMWh п $175ͬMWh). It addiƟonally purchases 2.4ථMWh from the grid at cost of $540 per year (including Įxed charges of $300, plus $240 for the energy consumed). This means that, at grid parity, the consumer pays a total of $820 per year for electricity when installing the solar PV system, higher than without it (Figureථ6.10b). A more accurate means of gauging the break-even point of solar PV is to consider ͞cost parity͟, which measures the point at which a household that installs a solar PV system incurs the same overall costs as it would if solely reliant on the grid. Using the example, 218

World Energy Outlook 2013 | Global Energy Trends

the cost of the PV system would have to drop to $160 (1.6ථMWh п $100ͬMWh), well below some current noƟons of grid parity, for it to make economic sense (Figureථ6.10c). This is equal to the variable cost that the PV system is displacing. “grid parity” and “cost parity” approaches

Dollars per MWh

Total cost: $700

(b)

Grid parity: $820

(c)

4

Cost parity: $700

250

Fixed system costs

5

Variable costs

6

Solar PV

7

200 $300 150

$300

100 50

$300

$280 $400

$240

$160

$240

Grid

PV

Grid

1 600 kWh

2 400 kWh

2 3

Figure 6.10 ‫ ٲ‬Indicative breakeven costs of residential solar PV using the (a)

1

8 9

Grid 4 000 kWh

PV

1 600 2 400 kWh kWh

10

There are several other factors that could inŇuence the compeƟƟveness of solar PV. For example, the calculaƟons would change if the household with the PV system did not consume all its generaƟon on site but sold the excess to the grid. In that case, for the PV generator to be fully compeƟƟve, the electricity would have to be sold at the actual wholesale market prices, i.e. the same price that other suppliers receive at that Ɵme. Any higher price for electricity sales, if Įxed by regulaƟon, would result in windfall proĮts for the PV generator. Also, if the renewables integraƟon costs were passed on to the seller, it would make compeƟƟveness harder to achieve. On the other side, potenƟal savings through reduced infrastructure needs could lower costs.

© OECD/IEA, 2013

Unit costs GeneraƟng costs for solar PV and wind power vary signiĮcantly across regions, according to local cost factors and the quality of the resources available (Figureථ6.11). The evoluƟon of these costs is largely determined by two factors: reducƟons in capital costs and technological advancements to harness more of the resource. Although increased deployment of wind and solar PV leads to prime sites becoming increasingly scarce within each region, which will tend to reduce capacity factors, at the global level this eīect is expected to be more than oīset by technological improvements and deployment in regions with untapped high-quality resources. Average capacity factors for wind onshore rise from 21% in 2012 to 26% in 2035, and for large-scale PV from 11% to 17% over the same period. Chapter 6 | Renewable energy outlook

219

11 12 13 14 15 16 17 18

Figure 6.11 ‫ ٲ‬Renewable electricity production costs relative to the wholesale prices for selected technologies and regions in the New Policies Scenario (a) Onshore wind Levelised costs of electricity

125

Wholesale price Uncertainty in price received by onshore wind producers

Dollars per MWh

150

100 75 50 25 2012 2020 2035 China

2012 2020 2035 United States

2012 2020 2035 European Union

Dollars per MWh

(b) Large-scale solar PV 250

Levelised costs of electricity

200

Wholesale price Uncertainty in price received by solar PV producers

150 100 50

2012 2020 2035 China

2012 2020 2035 United States

2012 2020 2035 European Union

© OECD/IEA, 2013

Notes: The cost of producƟon is the average levelised cost of electricity at deployed sites over the Outlook period, based on a weighted average cost of capital, assumed at 8% for OECD countries and 7% for nonOECD countries. Wholesale electricity prices are taken as the averages projected for respecƟve regions in the New Policies Scenario. In the mid-term, they include the recovery of investment costs for new capacity. The pricing methodology can be found at ǁǁǁ͘ǁŽƌůĚĞŶĞƌŐLJŽƵƚůŽŽŬ͘ŽƌŐ.

The global average investment cost of onshore wind was about $1ථ700ͬkW in 2012. Average costs for oīshore wind turbines are sƟll hard to quanƟfy, due to the small number of projects in operaƟon, but they are esƟmated to range from $3ථ000ͬkW to $4ථ500ͬkW. In the New Policies Scenario, the global average investment costs for onshore wind decrease by about 10%, partly due to learning eīects and economies of scale, as well as a shiŌ of new installaƟons towards non-OECD countries and related lower investment costs. Investment costs for oīshore wind decline by around one-third, with increased deployment. 220

World Energy Outlook 2013 | Global Energy Trends

Due to stepped up deployment and overcapacity in manufacturing, the price of solar PV systems dropped more than 40% between 2010 and end-2012. Demand for new capacity was around 30ථGW while producƟon capacity was about 55ථGW in 2012. Much of the growth in producƟon faciliƟes occurred in China, raising concerns that subsidies were enabling Chinese manufacturers to Ňood the European market with panels sold below cost. An agreement has been reached, prescribing a cap of 7ථGW per year and a minimum price for exports from China to the European Union and it is expected to run unƟl end 2015. Assuming that learning and economies of scale lead to an average cost decrease of about 20% each Ɵme capacity doubles (Frauenhofer ISE, 2012), about half of the price decrease over the last two years has been due to overcapacity. As this temporary situaƟon is resolved, market prices will tend to return to the long-term trend. Investments costs at the end of 2012 showed large regional diīerences. They ranged from some $1ථ800-5ථ500ͬkW for residenƟal rooŌop systems and $1ථ500-3ථ000ͬkW for large installaƟons, with China on the low side of the range, and the United States and Japan towards the higher side. In the New Policies Scenario, by the end of the Outlook period, the cost for both types declines by around 40%.

1

Bioenergy

8

ĞŵĂŶĚ Total global demand for bioenergy across all sectors increases from 1ථ300ථMtoe in 2011 to about 1ථ850ථMtoe in 2035, about two-thirds of the primary energy demand for natural gas today.6 The largest proporƟon of demand for bioenergy is in the buildings sector (including tradiƟonal biomass) throughout the Outlook period, though this declines in both absolute terms and share over Ɵme, largely as a result of relaƟvely high levels of demand in nonOECD countries. Demand for bioenergy in the power sector increases most in absolute terms from 2011 to 2035, by about 280 Mtoe, especially due to signiĮcant expansion in non-OECD countries such as China, India and Brazil. This growth is mainly driven by policies to reduce air polluƟon, boost producƟve use of domesƟc agricultural residues and speed deployment of renewables. Also driven by government support policies, demand for biofuels in the transport sector grows at the fastest rate over the Outlook period. It more than doubles in OECD countries and increases Įve-fold in non-OECD countries. Biofuels increase to more than 10% of total bioenergy demand (Figureථ6.12).

WƌŽĚƵĐƟŽŶĂŶĚdƌĂĚĞ To meet strong demand growth in the New Policies Scenario, the supply of all types of modern biomass will increase substanƟally, including biogas and municipal waste. Globally, the potenƟal supply of biomass exceeds the demand in 2035 by an order of magnitude, without compeƟng with food supply or displacing current forestry acƟviƟes, although land

2 3 4 5 6 7

9 10 11 12 13 14 15 16

© OECD/IEA, 2013

17 6.ഩ Global biomass use in the ͞other energy sector͟ in 2011 was reported at 65 Mtoe. In several cases, biomass use in biofuels production was understated, and it is estimated that there were some 50 Mtoe of unreported biomass use. If this amount would be included, it would result in an additional 200 Mtoe in 2035, or about 12% of global primary biomass demand.

Chapter 6 | Renewable energy outlook

221

18

use implicaƟons need to be carefully considered (IEA, 2012). 7 On the one hand, for some regions, it will be diĸcult for the domesƟc supply of various biomass feedstocks to keep pace with growing demand. For example, the European Union has taken strong measures to support the use of bioenergy in several sectors, including power and transportaƟon, but already imports large volumes of both biomass pellets and biofuels and will conƟnue to do so. India is another region that, despite having large supply potenƟal for many feedstocks, parƟcularly agricultural residues, struggles to ramp up the collecƟon of feedstocks to meet the strong growth in domesƟc demand for bioenergy, for both power sector applicaƟons and biofuels producƟon. The challenge to meet the growing demand domesƟcally will be especially diĸcult where markets already exist for waste products from agricultural and forestry acƟviƟes. On the other hand, a few regions have ample supplies to meet both domesƟc demand and internaƟonal demand. Brazil, Canada and the United States stand out in this group. Figure 6.12 ‫ ٲ‬World bioenergy use by sector in the New Policies Scenario 20%

World

40%

60%

80%

100%

2011

1 300 Mtoe

2035

1 850 Mtoe

2011

270 Mtoe

2035

510 Mtoe

2011

1 030 Mtoe

2035

1 340 Mtoe

OECD

Non-OECD

Buildings

Industry

Power

Transport

Other

Note: Buildings includes the use of tradiƟonal biomass.

© OECD/IEA, 2013

The demand for bioenergy for power generaƟon and heat increases from 136ථMtoe in 2011 to 420ථMtoe in 2035. Over 90%, of world demand is met from domesƟc resources throughout the Outlook period. To meet the remaining demand, some regions will increasingly turn to internaƟonal supplies of solid biomass for power generaƟon, most commonly in the form of biomass pellets. 8 In total, inter-regional trade of solid biomass for power generaƟon increases from a few percent of biomass consumpƟon to generate electricity to upwards of 8% by 2035. 7.ഩ This technical potential, based on conservative technological improvements, is an evaluation of available supply of forestry products, energy crops, and residues from forestry and agricultural activities. For more information, please see www.ǁŽƌůĚĞŶĞƌŐLJŽƵƚůŽŽŬ͘ŽƌŐ. 8.ഩ A processed product that has a relatively high energy density is fairly uniform and easier to transport than untreated biomass feedstocks. 222

World Energy Outlook 2013 | Global Energy Trends

The European Union is the largest importer of biomass for power generaƟon by 2035, imporƟng about 6.7ථMtoe. At current biomass pellet prices of $170 per tonne (Govan, 2012), the cost of these imports, largely coming from the United States, Canada and Russia, would reach almost $3ථbillion. Other regions may also become important players in this market, including countries in LaƟn America and Africa. In Japan, policy support pushes demand for biomass for power generaƟon and heat well beyond available domesƟc resources, driving up biomass imports to up to 4.4ථMtoe in 2035, coming mainly from Australia and the United States.
mboe/d

Figure 6.13 ‫ ٲ‬Biofuels demand and production in selected regions 1.5

1 2 3 4 5 6 7 8 9

Producon: Biodiesel

1.2

Ethanol

10

0.9

Demand: Biofuels

11

0.6

12

0.3

© OECD/IEA, 2013

2020 2035 2020 2035 European Union United States

2020 2035 Brazil

2020 2035 China

13

2020 2035 India

The internaƟonal market for biofuels increases from 0.2ථmboeͬd in 2012 to about 0.7ථmboeͬd in 2035, providing a broadly constant share of total biofuels demand over Ɵme. The European Union is the largest net importer of biofuels in 2035, with over 20% of its biofuels demand, aboutථ0.2ථmboeͬd (Figureථ6.13), met through imports from many diīerent countries, including Brazil, the United States and several countries in Asia and LaƟn America. These trade paƩerns re-emerge despite recent acƟon by the European Union to impose anƟ-dumping duƟes on biofuel imports from ArgenƟna, Indonesia and the United States, and the need for exporters to Europe to Įrst meet sustainability criteria, verifying reducƟons in greenhouse-gas emissions and demonstraƟng limited direct and indirect environmental impacts. Furthermore, a cap for convenƟonal biofuels of 6% (at the Ɵme of wriƟng) in the transport sector is under discussion in the European Union, which could have Chapter 6 | Renewable energy outlook

223

14 15 16 17 18

important impacts on the global picture. The United States is both a major importer and exporter of biofuels throughout the Outlook period, imporƟng sugarcane-based ethanol from Brazil to help meet rising targets for advanced biofuels under the Renewable Fuel Standard and exporƟng to the European Union to help meet blending targets there. The United States also conƟnues to export lower volumes of ethanol to Canada and Mexico. Brazil is the main supplier for internaƟonal biofuel markets, especially for fuel ethanol, and it is by far the largest exporter by the end of the Outlook period, providing about 0.2ථmboeͬd to the internaƟonal market by 2035, about 40% of global biofuels trade. China and India are both expected to increase their biofuels consumpƟon several Ɵmes over by 2035, making it diĸcult for domesƟc supply to keep up (Figureථ6.13). By 2035, both require some imports to meet demand. They are expected to come mainly from Brazil, but also from Indonesia and other countries in Asia. The assumed development of advanced biofuels at commercial scale aŌer 2020 aīects the biofuels market in several ways. First, it creates a single market for biomass feedstocks for the power and transport sectors. For some regions, this limits available supply for one or both of these sectors. For example, in India available supplies of residues become relaƟvely scarce by the end of the Outlook period due to demand from mulƟple sectors. The development of advanced biofuels also allows some regions to reduce their reliance on imports of biofuels, as they are able to tap alternaƟve feedstocks to produce biofuels. For example, the European Union is able to limit imports of biofuels aŌer 2025 due to a rise in domesƟcally produced advanced biofuels.

Investment

© OECD/IEA, 2013

CumulaƟve investment of $6.5ථtrillion (in year-2012 dollars) is required in renewable energy between 2013 and 2035 in the New Policies Scenario. This corresponds to $280ථbillion per year on average. Annual investments increase over the period, reaching almost $370ථbillion in 2035.9 Renewables for power generaƟon account for more than 95% of the total, with the remainder for biofuels. Projected investment for renewables in the power sector amounts to $6.2ථtrillion between 2013 and 2035 (Figureථ6.14). Renewables account for 62% of investment in new power plants over the projecƟon period, providing just over half of total capacity addiƟons. Wind power accounts for one-third of the total investment in renewables capacity, followed by hydropower (27%) and solar PV (23%). Investment for renewables in non-OECD countries are $3.3ථtrillion, higher than the $2.9 trillion required in OECD countries at. AddiƟonal global investment of $260ථbillion (4% of total grid infrastructure investment) is needed to upgrade transmission and distribuƟon networks to accommodate more renewables-based capacity. Investment to meet the expansion of biofuels supply amounts to $330ථbillion over 2013-2035, or $14ථbillion per year on average. ConvenƟonal producƟon of ethanol requires the bulk of the total (60%), followed by convenƟonal biodiesel (14%) and the remainder for advanced biofuels. OECD countries account for around 60% of total investment. 9.ഩ Investments for renewables used for heat are included in the buildings sector investments (see Chapter 7). 224

World Energy Outlook 2013 | Global Energy Trends

Figure 6.14 ‫ ٲ‬Cumulative investment in renewables-based power generation

Billion dollars (2012)

capacity, 2013-2035 2 100

Other OECD

1 800

European Union

2 3

United States

1 500

Other non-OECD

1 200

4

India China

900

5

600

6

300 Wind Bioenergy Hydro * Other includes geothermal, marine and solar CSP.

Solar PV

Other*

7

Subsidies

8

Renewable energy subsidies take a variety of forms, including blending mandates, quotas, porƞolio obligaƟons, tax credits and feed-in tariīs, which all oīer a higher return than market prices, to oīset higher costs. With schemes like feed-in tariīs, blending mandates or porƞolios and quota obligaƟons, this remuneraƟon is paid by the end-users (though some schemes, such as tax credits are funded from government budgets). Many forms of support mechanisms are speciĮc to electricity produced by renewables capacity installed in a parƟcular year, and have a Įxed duraƟon, typically twenty years. Subsidies for biofuels predominately take the form of blending mandates.

© OECD/IEA, 2013

Hydropower and geothermal, have long been economic in many locaƟons. Newer technologies, such as wind and solar, are oŌen an aƩracƟve opƟon for generaƟng electricity in remote, isolated areas with limited or no exisƟng grid infrastructure, but they require policy support to foster their deployment in most countries. The costs of generaƟon from onshore wind are geƫng closer to the average wholesale price level in many countries – and are already there in some, such as New ealand, Brazil, Ireland and parts of the United States. ReducƟons in producƟon costs for convenƟonal biofuels have not been as pronounced and these costs remain vulnerable to high feedstock prices and weather condiƟons. Outside Brazil and some parts of the United States, convenƟonal biofuels generally sƟll cost more than oil-based gasoline or diesel. Advanced biofuels remain even more costly, although there are promising signs that signiĮcant cost reducƟons in the producƟon process are on the horizon. In addiƟon to playing a crucial role in driving down the costs of renewable energy technologies, subsidies to renewables can have important co-beneĮts (Boxථ6.3). But support schemes for renewables need to be carefully designed to ensure their eĸciency and eīecƟveness. They should be predictable and transparent and, where possible, Chapter 6 | Renewable energy outlook

1

225

9 10 11 12 13 14 15 16 17 18

provide for compeƟƟon between technologies best suited to meet short- and long-term objecƟves. They need to be accompanied by ambiƟous, yet credible, targets and oīer support diīerenƟated according to the maturity of each technology. As cost reducƟons are achieved, the level of support provided for new installaƟons needs to decline to avoid unnecessary increases in the cost of energy services.  XOWLSOHEHQHÀWVRIUHQHZDEOHV Box 6.3 ‫ ٲ‬0

The contribuƟon of renewable energy to global energy needs has conƟnued to grow in recent years, sƟmulated by policy iniƟaƟves in an increasing number of countries. The beneĮts of renewables within a naƟonal energy porƞolio can be summarised as: „Energy security and diversity: renewable energy technologies can contribute to

energy security by providing more diversity in energy supply. They can also reduce the need for fossil fuels, in turn, reducing fuel import bills. „Environment: the deployment of renewables in the New Policies Scenario saves

some 4.1ථgigatonnes (Gt) of CO2 emissions in 2035 compared with the 2010 fuel mix at the same level of total generaƟon (IEA, 2012). Renewables also help reduce local air polluƟon and emissions of other pollutants, such as sulphur dioxide and nitrogen oxides. „Economic beneĮts: the development and deployment of renewables can form part

of comprehensive strategies aimed at more sustainable economic growth (oŌen called ͞green growth͟). Renewable energy has featured strongly in economic recovery packages put in place in response to the global economic downturn. „Energy access and aīordability: renewables can play an important role in providing

electricity access modern energy services to the 1.3ථbillion people currently without access to electricity and the 2.6ථbillion that sƟll rely on tradiƟonal use of biomass. Mini-grid and oī-grid soluƟons, including solar PV, are oŌen less costly than grid extension to rural areas (see Chapter 2).

© OECD/IEA, 2013

Based on a survey of established naƟonal policies, renewables are esƟmated to have received $101ථbillion in subsidies in 2012, 11% higher than 2011.10 This includes $82ථbillion to renewables for electricity generaƟon and $19ථbillion to biofuels for transport. The rise in 2012 was primarily due to the increase in solar PV capacity, increased generaƟon from capacity installed towards the end of 2011 and the increase of onshore wind capacity. The level of renewables subsidies is less than one-ĮŌh of the fossil-fuel consumpƟon subsidies in the same year (see Chapter 2). However, the geographical distribuƟon is very diīerent, with OECD countries paying about 85% of the renewables subsidies. Subsidies were most generous in the European Union ($57ථbillion) almost 60% of the total, the United States ($21ථbillion) and China ($7ථbillion). Ranked by generaƟng technology, subsidies for solar PV ($35ථbillion) were the highest, followed by wind ($26ථbillion) and bioenergy ($17ථbillion). 10.ഩ The subsidy estimates do not include integration costs or subsidies for renewable energy use for heat. See ǁǁǁ͘ǁŽƌůĚĞŶĞƌŐLJŽƵƚůŽŽŬ͘ŽƌŐ for the methodologies on how renewables subsidies and fossil-fuel consumption subsidies are calculated. 226

World Energy Outlook 2013 | Global Energy Trends

Subsidies to biofuels declined by almost 20% in 2012 from the previous year. This is partly due to reform of taxaƟon levels (or tax incenƟves) in some of the main subsidising regions, notably the United States and Brazil. In the United States, the US Congress did not extend the ethanol import tariī nor the producƟon tax credit for ethanol, which had been in place for decades. In Brazil, subsidies to biofuels fell by more than half due to lower ethanol supply, following a reducƟon in the blending mandate, and a decrease in the tax preference provided to ethanol, when gasoline taxes were reduced in the middle of the year. China, too, drasƟcally lowered subsidies for ethanol from $155ͬtonne to $80ͬtonne and, although biofuels use increased, the total subsidy level declined.

1

Figure 6.15 ‫ ٲ‬Global renewable energy subsidies by source in the New

5

Billion dollars (2012)

Policies Scenario 250

3 4

6

Biofuels Electricity:

7

200

Other

150

Concentrang solar power Wind offshore Wind onshore

100

Bioenergy

8 9

Solar PV

50

10 2008 2010 2012

2015 2020 2025 2030

2035

11

Notes: Other includes geothermal, marine and small hydro.

The New Policies Scenario projects an almost six-fold increase in electricity generaƟon from non-hydro renewables and a tripling of the use of biofuels. Subsidies to renewable energy amount to over $220ථbillion per year in 2035, aŌer peaking just above $240ථbillion around 2030 (Figureථ6.15). From 2013 to 2035, cumulaƟve subsidies to renewables amount to $4.7ථtrillion, or around 0.15% of cumulaƟve global GDP. These esƟmates are calculated by taking the diīerence between the levelised cost of electricity generated by the renewable energy technology and the regional wholesale electricity price, mulƟplied by the amount of generaƟon. For biofuels, they are calculated by mulƟplying the volumes consumed by the diīerence between their cost and the reference price of the comparable oil-based products.

© OECD/IEA, 2013

2

In total, annual subsidies for renewables for power generaƟon reach $177ථbillion in 2035. They peak around 2030 and then decline, thanks to increasing wholesale power prices, the decreasing unit costs of most renewable energy technologies, and because older installaƟons are gradually reƟred, meaning that newer and cheaper units make up a larger share of the installed capacity. Subsidies for onshore wind power peak just aŌer 2020, earlier than those for any other technology, reŇecƟng the increasing compeƟƟveness Chapter 6 | Renewable energy outlook

227

12 13 14 15 16 17 18

of this technology. Subsidies for biofuels conƟnue to increase unƟl 2030 to $50ථbillion, reaching $45ථbillion in 2035. The bulk of this goes to convenƟonal biofuels, which remain uncompeƟƟve in most locaƟons with convenƟonal gasoline or diesel. Over the projecƟon period, onshore wind becomes compeƟƟve in more and more regions, generaƟng over 33ථ300ථTWh cumulaƟvely, supported by subsidies of some $620ථbillion (Figureථ6.16), corresponding to a low level of subsidy per unit of output ($19ͬMWh on average over the Outlook period, including non-subsidised generaƟon). Solar PV requires $1ථ600 billion of cumulaƟve subsidies and generates almost 12ථ200ථTWh, meaning a higher average unit subsidy of $131ͬMWh. Where support policies are commiƩed for many years (typically twenty years for feed-in tariīs), subsidies for older capacity conƟnue to be paid, even aŌer new projects reach compeƟƟveness. For bioenergy, the unit subsidy remains broadly constant through 2035, as costs are not expected to decline signiĮcantly. For this reason, and due to a more than three-fold increase in bioenergy generaƟon, subsidies to bioenergy become the second-largest of all by 2035, behind solar PV. Figure 6.16 ‫ ٲ‬Global subsidies for renewable electricity generation and 1 800

36 000

1 500

30 000

1 200

24 000

900

18 000

600

12 000

300

6 000 Solar PV

Wind onshore

Wind Bioenergy offshore

CSP*

TWh

Billion dollars (2012)

generation by source in the New Policies Scenario, 2013-2035 Subsidy Electricity generaon (right axis): Unsupported Supported by subsidy

Others

© OECD/IEA, 2013

* ConcentraƟng solar power.

Similar to the global trend, total subsidies to electricity from renewables peak over the projecƟon period in the European Union and in China (Figureථ6.17). In the European Union, subsidies level oī in 2020 at $60 billion per year, before declining to about half that level by 2035, as the subsidies to some 52ථGW of PV solar added in the last three years come to an end and wholesale prices increase. The peak in China reaches about $35 billion per year around 2030 and then falls to below $30ථbillion in 2035. The bulk of the increase is aƩributable to bioenergy and solarථPV. The share of wind decreases from 37% in 2012 to 26% in 2035. In the United States, subsidies increase from $13ථbillion in 2012 to just above $30ථbillion in 2035, while generaƟon from non-hydro renewables almost quadruples. Subsidies to onshore wind decrease over the Outlook period, as the technology gains in compeƟƟveness, while subsidies to bioenergy increase strongly, due to a three-fold increase in generaƟon and the replacement of a large amount of aging capacity.

228

World Energy Outlook 2013 | Global Energy Trends

Figure 6.17 ‫ ٲ‬Renewables-based generation subsidies by source and

1

selected region in the New Policies Scenario

Billion dollars

(a)

United States

35 historical

2

Other

projecons

30

Concentrang solar power

25

Wind offshore Wind onshore

20

Bioenergy Solar PV

15

3 4 5

10

6

5 2008

2012

2015

2020

2025

2030

7

2035

Billion dollars

(b) European Union 70 historical

8

Other

projecons

60

Concentrang solar power

50

Wind offshore Wind onshore

40

Bioenergy

11

20 10

12

2008

2012

2015

2020

2025

(c) Billion dollars

10

Solar PV

30

2030

2035

13

China

40 35

9

historical

Other

projecons

Concentrang solar power

30

Wind offshore

25

Wind onshore Bioenergy

20

14 15 16

Solar PV

15

© OECD/IEA, 2013

10

17

5 2008

2012

2015

2020

2025

2030

18

2035

Note: Other includes geothermal, marine and small hydro.

Chapter 6 | Renewable energy outlook

229

© OECD/IEA, 2013

Chapter 7 (QHUJ\HIÀFLHQF\RXWORRN On track for more with less? +LJKOLJKWV

x A government-led renewed focus on energy eĸciency, at a Ɵme of higher energy prices, has accelerated the previously slow rate of improvement in global energy intensity. The amount of energy used to produce a unit of GDP declined by 1.5% in 2012, compared with an average annual decline of just 0.4% between 2000 and 2010. The biggest improvements in 2012 were in countries with energy intensiƟes above the global average, including China and Russia. China’s energy intensity has gone from almost four Ɵmes the world average in 1990, to less than twice the world average now, driven by signiĮcant improvements in industry.

x MoƟvated both by compeƟƟveness consideraƟons and environmental factors, major new energy eĸciency policies are being implemented or discussed, including the EU Energy Eĸciency DirecƟve, strengthened appliance standards in the United States and iniƟaƟves under the US-China Climate Change Working Group to enhance truck eĸciency standards and eĸciency in buildings in both countries. In the New Policies Scenario, the implementaƟon of policies for eĸciency, which are under discussion, leads to savings of 910ථMtoe in 2035 (compared with the Current Policies Scenario), which is just over half of the current energy use of the European Union.

x Industry accounts for 37% and buildings for 26% of eĸciency-related primary energy savings in 2035 in the New Policies Scenario. In both sectors, the bulk of the savings are made in the use of electricity, led by eĸciency improvements in electric motor systems, stricter standards for appliances, and more eĸcient lighƟng. Improved fueleconomy standards in transport lead to oil savings of around 5ථmbͬd in 2035, or 31% of the total primary energy savings. Improvements in the average eĸciency of fossil fuel-Įred power plants, parƟcularly in China, India and the United States, account for most of the remainder.

x Subsidies to energy consumpƟon are a major barrier to investments in energy eĸciency. Depending on the region, the payback period can be extended up to nine Ɵmes. For example, in the Middle East, due to low gasoline prices, an investment in a hybrid car is recovered through lower fuel costs only aŌer almost eighteen years, compared with four years in Europe.

© OECD/IEA, 2013

x The New Policies Scenario gives rise to an addiƟonal $3.4ථtrillion of investment in energy eĸciency through to 2035 (on top of the $4.7ථtrillion required in the Current Policies Scenario). These addiƟonal investments generate cumulaƟve savings in energy expenditures of $6.1ථtrillion. Household energy bills are reduced by $2.6ථtrillion through energy eĸciency. These savings free up disposable income, which leads to an increase in the use of non-energy goods. Chapter 7 | Energy efficiency outlook

231

IntroducƟon Although not always as visible as supply-side opƟons, energy eĸciency is an essenƟal component of a sustainable energy future. Policies to improve the eĸciency of energy use can start delivering beneĮts fairly quickly, including by improving energy security and industrial compeƟƟveness, cuƫng household energy bills and reducing problems linked to local air polluƟon and climate change. In contrast to supply-side opƟons, energy eĸciency opƟons are oŌen obscured by the fact that eĸciency is rarely traded or priced. Furthermore, improving eĸciency involves a wide range of acƟons aīecƟng a variety of energy services across diīerent sectors – including buildings, industry and transport – so the overall achievement is oŌen diĸcult to quanƟfy. Energy eĸciency was the ͞fuel͟ of focus in the 2012 ediƟon of the World Energy Outlook (WEO) (Boxථ7.1). The report found that policies and measures that are enacted or are currently under discussion will fall well short of tapping the full economic potenƟal of energy eĸciency by 2035 (Figureථ7.1). In order to raise the visibility of energy eĸciency and to put it on the same fooƟng as its supply-side alternaƟves, we have decided to dedicate a chapter of each year’s WEO to energy eĸciency, along with the chapters on the primary fuels and electricity.1 This decision is designed to encourage and support the increasing aƩenƟon that is now being directed to energy eĸciency policy in all parts of the world, not least as a means of reducing costs and improving compeƟƟveness (see Chapterථ8). Figure 7.1 ‫ ٲ‬3  URSRUWLRQRIORQJWHUPHFRQRPLFHQHUJ\HIÀFLHQF\SRWHQWLDO DFKLHYHGLQWKH1HZ3ROLFLHV6FHQDULR 100%

Unrealised energy efficiency potenal

80%

Realised energy efficiency potenal

60% 40% 20%

Industry

Transport

Power generaon

Buildings

Source: IEA (2012a).

© OECD/IEA, 2013

This chapter Įrst considers recent energy eĸciency trends by region and sector and discusses recent policy developments. It then focuses on the impact of energy eĸciency policies yet to be implemented or under discussion, as assumed in the New Policies 1.ഩ For similar reasons, alongside the medium-term market reports for different fuels, the IEA has recently published the first edition of the Energy Efficiency Market Report (IEA, 2013a). 232

World Energy Outlook 2013 | Global Energy Trends

Scenario. Energy eĸciency is examined by sector, fuel and region. The analysis quanƟĮes the avoided energy use due to energy eĸciency in the New Policies Scenario compared with the Current Policies Scenario. Further, the chapter discusses energy eĸciency investment needs and describes the mulƟple beneĮts the energy eĸciency policies under discussion would provide, including macroeconomic beneĮts, savings in import bills and reduced levels of local air polluƟon and carbon-dioxide (CO2) emissions. Box 7.1 ‫ ٲ‬7 KH(IÀFLHQW:RUOG6FHQDULR²WDFNOLQJFRPSHWLWLYHQHVVHQHUJ\

1 2 3 4

VHFXULW\DQGFOLPDWHFKDQJHVLPXOWDQHRXVO\

To quanƟfy the implicaƟons for energy markets, the economy and the environment of undertaking all economically viable energy eĸciency investments, the WEO-2012 (IEA, 2012a) presented the Eĸcient World Scenario, seƫng out the policies needed to overcome the various barriers to the comprehensive adopƟon of energy eĸciency measures. This scenario shows that the policies under discussion, included in the New Policies Scenario, achieve only one-third of the economic potenƟal of energy eĸciency (Figureථ7.1). The Eĸcient World Scenario provides a blueprint that idenƟĮes the policies and measures required in each sector to unlock this full potenƟal, along with an esƟmate of the required investment. Those investments pay back well before the end of the technical lifeƟme of the energy capital stock. In the Eĸcient World Scenario, growth in primary energy demand through to 2035 is halved (in net terms, i.e. aŌer taking the rebound eīect into account), relaƟve to the New Policies Scenario. The vast majority of savings relaƟve to the New Policies Scenario are achieved by end-users: more than 40% in buildings,2 23% in industry and 21% in transport. The Eĸcient World Scenario leads to a more eĸcient allocaƟon of resources and delivers economy-wide beneĮts. The global economy generates an addiƟonal $18ථtrillion in cumulaƟve output by 2035, corresponding to the combined gross domesƟc product (GDP) of the United States, Canada, Mexico and Chile in 2011. The energy security of imporƟng naƟons increases, while exporters are able to sell volumes, which would otherwise have gone to internal consumpƟon. Along with reduced local polluƟon, energy-related CO2 emissions peak before 2020 and decline thereaŌer, a trajectory consistent with a long-term temperature increase of 3ථΣC. This is a signiĮcant improvement compared to the New Policies Scenario, though sƟll falling short, in isolaƟon, of limiƟng the global temperature increase to 2ථΣC. The aim of the Eĸcient World Scenario was to highlight the very large potenƟal for energy eĸciency. As this does not change on a yearly basis, the Eĸcient World Scenario has not been updated for this ediƟon of the Outlook.

5 6 7 8 9 10 11 12 13 14 15 16

© OECD/IEA, 2013

17 2.ഩBuildings comprise the residential and services sector, including energy consumption for space and water heating, cooking, lighting, appliances and cooling.

Chapter 7 | Energy efficiency outlook

233

18

Current status of energy eĸciency Recent progress Global energy intensity, measured as the amount of energy required to produce a unit of GDP, declined by 1.5% in 2012, which was similar to the improvement in 2011.3 A renewed policy focus on energy eĸciency and higher energy prices have together accelerated the previous slow pace of energy intensity improvement (Boxථ7.2). The decade to 2010 had seen an average annual improvement of only 0.4% per year with energy intensity worsening in 2009 and 2010, partly because of colder-than-usual winters and the economic recession. Box 7.2 ‫ ( ٲ‬QHUJ\HIÀFLHQF\HQHUJ\LQWHQVLW\DQGHQHUJ\VDYLQJV4 Energy intensity, in general deĮned as primary energy demand per unit of economic output, is not a good indicator of energy eĸciency as it is inŇuenced by the structure of an economy and climaƟc condiƟons (IEA, 2008 and 2012a). For analysis of past trends, energy intensity is nevertheless oŌen used as a proxy for energy eĸciency in the absence of more detailed data. This chapter discusses trends in energy eĸciency in the New Policies Scenario. Energy savings are stated by reference to the Current Policies Scenario, which does not aƩempt to capture the eīect of policies under discussion but not yet adopted. These savings can arise in end-use sectors (i.e. transport, buildings, industry and agriculture) and in supply (i.e. power generaƟon, oil and gas extracƟon and reĮneries).

© OECD/IEA, 2013

Energy savings can be grouped in three broad categories: reducƟons in the demand for energy services; savings due to fuel and technology switching; and savings due to energy eĸciency improvements. Changing end-user prices lead to changes in the demand for energy services, reducing the demand for Įnal energy consumpƟon. Fuel and technology switching, for example switching from gasoline cars to electric cars, adopƟng heat pumps for space and water heaƟng or producing steel in electric arc furnaces instead of basic oxygen furnaces can reduce Įnal consumpƟon. Energy eĸciency savings, strictly deĮned, are diīerent: they provide the same energy service while consuming less energy. Such energy eĸciency improvements may be provided by adopƟng more eĸcient technologies, including, for example, beƩer insulaƟon of a buildings shell, or by improving system eĸciency through energy management systems and process control. In order to disƟnguish these three diīerent eīects, we employ a decomposiƟon analysis based on our technology-rich World Energy Model results. 4

3.ഩථWhile energy intensity is presented here using GDP at market exchange rate (MER), it can also be expressed in terms of purchasing power parity (PPP). For a discussion on the use of GDP at MER or PPP see also IEAͬWorld Bank (2013). 4.ഩථFor more informaƟon on the decomposiƟon analysis and the World Energy Model, see ǁǁǁ͘ǁŽƌůĚĞŶĞƌŐLJŽƵƚůŽŽŬ͘ŽƌŐͬǁĞŽŵŽĚĞůͬĚŽĐƵŵĞŶƚĂƟŽŶͬ͘ 234

World Energy Outlook 2013 | Global Energy Trends

1

Figure 7.2 ‫ ٲ‬3  ULPDU\HQHUJ\LQWHQVLW\OHYHOVDQGWUHQGVLQVHOHFWHGUHJLRQV

4

toe per thousand dollars of GDP ($2012, MER)

Energy intensity in 2012 declined, in relaƟve terms, mostly in countries where its level was above the global average, leaving greater remaining potenƟal for improvement (Figureථ7.2). Countries with the highest energy intensiƟes oŌen have considerable fossil fuel resources and oŌen subsidise fossil fuel consumpƟon (IEAͬWorld Bank, 2013). On the contrary, those countries with the lowest energy intensiƟes among the world’s toptwenty energy consumers are all characterised by high energy prices and strict eĸciency legislaƟon, e.g. Japan and members of the European Union.

0.6

6%

0.4

4%

0.2

2%

Energy intensity

3

5

Change from 2011 (right axis)

0%

6 7

-2%

8

-4% -6% India

© OECD/IEA, 2013

2

Russia

China

9

Middle ASEAN United European Japan East States Union

Notes: GDP is expressed in year-2012 dollars in market exchange rate (MER) terms; toe с tonnes of oil equivalent.

10

In 2012, the United States made the biggest relaƟve improvement in energy intensity, as a result of eĸciency gains in industry and services, together with fuel switching in power generaƟon to natural gas. The second-biggest improvement was in Russia due primarily to lower energy use per unit of output in industry and services. Russia was closely followed by China, where eĸciency improvements in industry (Boxථ7.3) and higher output from hydropower and other renewables led to a 4% improvement in energy intensity in 2012.5 From being four Ɵmes higher than the global average in 1990, China’s energy intensity is now less than twice the global average. In the Middle East, in contrast to other regions, energy intensity increased in 2012, mainly due to increased acƟvity in energy-intensive industry (e.g. petrochemicals) and rapid energy demand growth in buildings and transport.

11

At its simplest, assessment of the improvement in energy intensity simply involves measuring the reducƟon in energy use per unit of GDP that has been achieved over a given period. On this basis, energy intensity across almost all regions improved over the last two decades. Between 1990 and 2000, global energy intensity (disregarding changes in the regional make-up of regional GDP) improved by 1.4% per year, and 1.8%ථper year over the period since 2000 (Figureථ7.3).

15

5.ഩ The physical energy content method, applied by the IEA, uses the physical energy content of the primary energy source as the primary energy equivalent. Accordingly, the conversion efficiency for hydro and wind power is 100%.

Chapter 7 | Energy efficiency outlook



12 13 14

16 17 18

Box 7.3 ‫ ( ٲ‬QHUJ\HIÀFLHQF\GRHVGHOLYHU6 There has been a long-standing debate on whether energy eĸciency really can deliver long-term energy savings, mostly due to diīerent esƟmates of the rebound eīect and the extent to which demand savings will inŇuence energy prices (IEA, 2012a; Sorrell, 2007; Frondel, RiƩer and Vance, 2012). A common percepƟon is that liƩle progress has been made in improving energy eĸciency. However as measurement of energy eĸciency improves, it is possible to show evidence that it really is making a diīerence. Here are three examples of recent policies and their impact: „ During its 11th and 12th Five-zear Plans, China introduced wide-ranging policies

to reduce energy use in industry, which accounts for about half of its total Įnal energy consumpƟon. China’s industrial energy consumpƟon is dominated by the steel and cement sector, which previously consumed much more energy per unit of output than the OECD average. Measures that have reduced industrial energy intensity have included energy performance standards; targets for annual energy consumpƟon for the biggest companies; the closure of small and outdated plants; energy audits and the introducƟon of energy management systems; and Įnancial incenƟves. Oĸcial reports and independent analysis esƟmate that China saved 105ථ million tonnes of oil equivalent (Mtoe) in its industrial energy demand over four years (2006-2009) through its Top-1ථ000 Programme, an amount equivalent to the enƟre annual energy demand of Poland (Jing, et al., 2012). The energy intensity of China’s cement industry is now comparable with the OECD average. „ Japan’s Top-Runner Programme, set the Įrst fuel-economy standards in the

transport sector in 1999. The goal was to reduce speciĮc fuel consumpƟon of new passenger light-duty vehicles by about 23% within ĮŌeen years. The target was achieved in 2005, Įve years ahead of schedule. The Top-Runner Programme is esƟmated to have saved around 0.2ථmillion barrels per day in oil consumpƟon in 2010 and thus cut annual fuel bills by almost $6ථbillion ($2012) in the same year (GWPH, 2013). „ The German
© OECD/IEA, 2013

loans and other subsidies to incenƟvise renovaƟons in exisƟng buildings to improve their energy eĸciency (75% of German homes were built before any energy eĸciency regulaƟons came into force). RenovaƟons Įnanced through the Bank improved energy performance in houses and reduced their average energy consumpƟon by 30% in 2011. It has been independently esƟmated that the
6.ഩIncreased energy efficiency can lead beneficiaries to spend part of the revenue saved on new or additional energy services, so offsetting part of the original efficiency savings. This is called rebound effect and is taken into account in the World Energy Model projections. 236

World Energy Outlook 2013 | Global Energy Trends

But, at the same Ɵme, countries with higher energy demand per unit of GDP were increasing their share of global economic output: the shares of diīerent regions in global output were shiŌing. Taking this into account, there has been a signiĮcant slowdown in the global rate of energy intensity improvement. Global energy intensity on this basis declined by 1.3% per year on average in the 1990s, but the decline dropped to 0.4% per year during the 2000s. The case of China illustrates this eīect. China’s economy is signiĮcantly more energy intensive than the world average. As its weight in the global economy increases, so the global average improvement in energy intensity is dampened (Figureථ7.3). At the start of the 1990s, this eīect was oīset by the collapse of the Soviet Union and the consequent decline in its economic acƟvity.  QQXDOUHODWLYHFKDQJHLQJOREDOSULPDU\HQHUJ\LQWHQVLW\E\ Figure 7.3 ‫ ٲ‬$ GULYHU 2%

Change in energy intensity excluding regional economic shis

1% 0%

Change in energy intensity from change in regional contribuon to global GDP

-1% -2%

Change in overall energy intensity

-3%

1995

2000

2005

Recent policy developments

© OECD/IEA, 2013

3 4 5 6 7 8 9

11

2010 2012

Notes: Changes in energy intensity are split into two drivers – energy intensity on a regional level and regional GDP share – using a rolling decomposiƟon analysis. Energy intensity is measured using GDP at market exchange rate in year-2012 dollars.

Improvements in energy eĸciency over the past two years have been accompanied by encouraging signs of increasing acƟon on the policy front in many regions (Tableථ7.1). In July 2013, China and the United States draŌed a co-operaƟon plan, pledging to make heavy-duty vehicles and coal-Įred power plants more eĸcient. Further, the US administraƟon’s plan on climate acƟon, announced in June 2013, has strong energy eĸciency components. The plan includes: the imposiƟon of carbon polluƟon standards on new and exisƟng power plants (which would lead to increased average fossil-fuel power plant eĸciency); strengthening fuel-economy standards for heavy-duty vehicles beyond 2018; and Ɵghtening eĸciency standards for electric appliances. In 2012, Canada extended fuel-economy standards for cars to 2025, introduced stringent performance standards on new power plants (banning the construcƟon of coal-Įred power staƟons, unless equipped with carbon capture and storage technology) and introduced Minimum Energy Performance Standards (MEPS) for several products, including lighƟng and water heaƟng. Chapter 7 | Energy efficiency outlook

2

10

-4% 1990

1

237

12 13 14 15 16 17 18

Among others in OECD, the European Union’s Energy Eĸciency DirecƟve, which came into force in December 2012, must be implemented by member states by mid-2014. Several regulaƟons for appliances, such as computers and vacuum cleaners, have been implemented under the EU Ecodesign DirecƟve. In Australia, the government has taken further steps as part of the Clean Energy Future package to exploit the remaining energy eĸciency potenƟal and Įnancial support will be provided and research promoted through the Clean Technology Program and the Clean Energy Finance CorporaƟon. In industry, the Energy Eĸciency OpportuniƟes programme has been opened up to medium-size energy consumers. Discussion has also begun on seƫng eĸciency standards for light-duty vehicles. Japan added windows and heat-insulaƟng materials to the Top-Runner Programme in May 2013, which is expected to promote technology innovaƟon. Also in Japan, an evaluaƟon is to be carried out of industrial consumer’s eīorts to reduce electricity use during peak hours. AcƟon on energy eĸciency has not been conĮned to OECD countries. The world’s biggest energy consumer, China, introduced an extensive set of eĸciency goals and measures at the start of its 12th Five-zear Plan in 2011 with the objecƟve of reducing energy intensity 16% by 2015 compared with 2010. Over the past year, some concrete policy acƟons have been taken to realise the target. The NaƟonal Development and Reform Commission announced more market-oriented pricing for oil products and a price increase for natural gas for businesses. Furthermore, Beijing plans to eliminate large coal-Įred boilers from the city centre by the end of 2015. In India, the Energy ConservaƟon in Building Code has become mandatory for large commercial and residenƟal buildings in several states, with the objecƟve of reducing energy consumpƟon from lighƟng and hot water systems, and the MEPS for air condiƟoners have been Ɵghtened. Southeast Asian countries are also increasingly turning aƩenƟon to energy eĸciency, but subsidies remain commonplace across the region (IEA, 2013b). In Singapore, the Energy ConservaƟon Act has been implemented, requiring a co-ordinated industrial approach to energy eĸciency and, in Malaysia, a long-term NaƟonal Energy Eĸciency Master Plan has been adopted.

© OECD/IEA, 2013

Brazil has joined other countries in adopƟng incenƟves to increase energy eĸciency in the transport sector. The Inovar-Auto programme encourages technology innovaƟon by requiring car manufacturers to increase the eĸciency of cars up to 2017 in order to qualify for tax breaks. Its eīecƟve implementaƟon is expected to increase the eĸciency of lightduty vehicles by at least 12% by 2017 (see Chapterථ10). The Inovar-Auto programme is seen as a Įrst step towards the establishment of mandatory targets. The Middle East saw some intensiĮed acƟon on limiƟng the spiralling electricity consumpƟon from air condiƟoners, which account for roughly half of electricity demand in that region. Saudi Arabia strengthened MEPS for air condiƟoners, while the United Arab Emirates introduced such regulaƟon for the Įrst Ɵme.

238

World Energy Outlook 2013 | Global Energy Trends

Table 7.1 ‫ ٲ‬6 HOHFWHGHQHUJ\HIÀFLHQF\SROLFLHVDQQRXQFHGRULQWURGXFHGLQ DQG New policy measures

Sector United States

Canada

Japan

European Union

Announcement of carbon polluƟon standards for new and exisƟng power plants.

Buildingsͬ industry

Proposal of the Energy Savings and Industrial CompeƟƟveness Act of 2013 to: strengthen building codes; create a Įnancing iniƟaƟve; and incenƟvise the applicaƟon of eĸcient motors. Tighten eĸciency standards for appliances.

3

Transport

IntenƟon announced to increase fuel-economy standards for heavy-duty vehicles beyond 2018.

4

Transport

Proposed extension of emissions standards for passenger and commercial lightduty vehicles beyond 2016; implementaƟon of emissions standards for heavy-duty vehicles.

5

Buildings

Increased stringency of MEPS for several products, including lighƟng, water heaƟng, air condiƟoners and appliances.

6

Power

IntroducƟon of performance standards requiring new power staƟons not to exceed 420ථtonnesථof CO2 per gigawaƩ-hour.

Buildingsͬ industry

Extension of the Top-Runner Programme to windows and insulaƟng materials.

7

EvaluaƟon of industrial consumer eīorts to reduce electricity use during peak hours.

8

ImplementaƟon of regulaƟons for vacuum cleaners and computers within the framework of the Ecodesign DirecƟve.

Transport

Agreement on a fuel-economy standard for new cars of 95 grammes of CO2ͬkm by 2020.

Australia

Industry

Clean Technology Program invests $1.2 billion to improve energy eĸciency and support research and extension of the Energy Eĸciency OpportuniƟes programme. Establishment of the Clean Energy Finance CorporaƟon endowed with $10 billion fund to invest in clean energy including energy eĸciency.

China

General

Energy price reform (more frequent adjustments in oil product prices and an increase in natural gas price by 15% for non-residenƟal users).

Buildings

IntroducƟon of energy standards for new buildings and the refurbishment of exisƟng dwellings.

Buildings

Southeast Industry Asia

9 10 11 12

More stringent MEPS for air condiƟoners. Energy ConservaƟon in Building Code (ECBC) made mandatory in eight states. It applies, among other things, to the building envelope, lighƟng and hot water. Singapore: Energy ConservaƟon Act came into force requiring reporƟng on energy use, appoinƟng energy managers and elaboraƟng eĸciency improvement plans. Malaysia: NaƟonal Energy Eĸciency Master Plan established an overall long-term plan for eĸciency, with a goal to reduce electricity consumpƟon by 10% in 2020.

© OECD/IEA, 2013

2

Power

Buildings

India

1

13 14 15

Brazil

Transport

Inovar-Auto programme approved requiring car manufacturers to produce more eĸcient vehicles to qualify for a tax discount.

Africa

Buildings

Economic Community of West African States (ECOWAS): phase-out of incandescent lighƟng by 2020, adopƟon of appliance standards and labels by 2014 and development of region-wide standards for buildings.

Industry

South Africa: Manufacturing CompeƟƟveness Programme of $0.6ථbillion with one aim being to upgrade current producƟon faciliƟes.

17

Saudi Arabia: more stringent MEPS for air condiƟoners.

18

Middle East

Buildings

United Arab Emirates: introducƟon of MEPS for air condiƟoners and mandatory energy labelling scheme for all domesƟc appliances.

Chapter 7 | Energy efficiency outlook



16

Recent sectoral trends Diverging energy intensity developments by sector are apparent at a regional level (Figureථ7.4). Global industrial energy intensity decreased by only 3% between 2005 and 2012. In China it decreased by one-quarter, while in the Middle East it increased by oneĮŌh.7 In the United States, energy intensity in industry decreased only slightly over the past seven years as eĸciency improvements were almost fully oīset by an increase in oil and gas producƟon and increased acƟvity in the chemicals industry that shiŌed the economy, to some extent, to more energy-intensive sectors. The European Union, on the other hand, saw a decline of about 15% in its industrial energy intensity, parƟally linked to the declining share of energy-intensive industry, such as iron and steel, in total industrial output. Energy intensity levels in Japan’s industry sector decreased by about 9% from 2005 to 2012, helped by structural changes in the economy away from energy-intensive sectors, including metals and paper. Figure 7.4 ‫ ( ٲ‬QHUJ\LQWHQVLW\FKDQJHE\VHFWRUDQGUHJLRQ Industry

Passenger light-duty vehicles

Residenal

United States European Union Japan China India World -28%

-14%

0%

-20%

0%

20%

-20%

-10%

0%

DecomposiƟon analysis for China shows that structural change from 2005-2010 played only a minor part in the substanƟal improvement in its industrial energy intensity in those years, meaning that the bulk of the improvement can be aƩributed to energy eĸciency gains (Hasanbeigi, et al., 2013). The share of energy-intensive industries in total industrial value added did not change signiĮcantly during the 11th Five-zear Plan, due to strong growth in cement and steel producƟon. Driven by ambiƟous energy eĸciency policies, including the ten key projects and Top-1ථ000 Energy-consuming Enterprises Programme, eĸciency improvements were strongest in the cement and paper industries.

© OECD/IEA, 2013

In the residenƟal sector, which accounts for one-quarter of global Įnal energy demand and 73% of total buildings energy demand, energy intensity fell by 20% over the past seven years, but regional trends diverge.8 In emerging economies, such as China, savings due 7.ഩ Industrial energy intensity is defined as energy consumed per unit of industrial value added. 8.ഩ Residential energy intensity is defined as modern residential fuel use per square metre and capita, calculated only where the population has access to modern energy. 

World Energy Outlook 2013 | Global Energy Trends

to improvements in energy eĸciency are more than oīset by an increase in consumpƟon driven by higher wealth and living standards. In OECD countries, energy intensiƟes in households decreased over the period, mainly as a consequence of eĸciency measures. In countries where parts of the populaƟon lack access to electricity, such as India, the intensity of energy use decreased markedly, as households gaining access to modern energy supply normally consume relaƟvely liƩle compared with the country’s average consumpƟon of modern energy per capita and per square metre of Ňoor space. This eīect strongly impacts the global intensity trend; therefore a decrease in energy intensity is not enƟrely driven by eĸciency improvements. In road transport, diverging regional trends can also be observed from 2005: in Japan and the European Union, high fuel prices, fuel standards and the sluggish economy led individuals to buy smaller cars, which reduced fuel intensity; while in India fuel intensity stayed roughly constant. 9

1 2 3 4 5 6

The outlook for energy eĸciency

7

In the New Policies Scenario, energy demand to 2035 increases by one-third, compared with almost 45% in the Current Policies Scenario, saving 1ථ260ථMtoe in 2035. Eĸciency accounts for almost three-quarters, or 910ථMtoe, of the energy savings in 2035, reŇecƟng the policies and measures already being discussed (Figureථ7.5). The bulk of the savings occur in end-use sectors, with a much smaller share achieved in energy supply and transformaƟon. The amount of energy required to generate a unit of GDP is reduced by 37% compared with today. Global energy intensity improvements average 1.9% per year, compared with 1.6% in the Current Policies Scenario (and 2.5% in the 450ථScenario).

10

Figure 7.5 ‫& ٲ‬  KDQJHLQJOREDOSULPDU\HQHUJ\GHPDQGE\FDWHJRU\LQWKH

11

8 9

1HZ3ROLFLHV6FHQDULRUHODWLYHWRWKH&XUUHQW3ROLFLHV6FHQDULR

12 13 14 15 16

© OECD/IEA, 2013

17 9.ഩ Transport energy intensity is defined as on-road fuel consumption in passenger light-duty vehicles in litres per 100 kilometres.

Chapter 7 | Energy efficiency outlook

241

18

Apart from greater eĸciency, other measures that contribute to reducing energy demand include fuel switching and reduced demand for energy services.10 Fuel and technology switching is parƟcularly concentrated in the power sector being achieved by increasing the share of generaƟon from gas-Įred power plants, solar photovoltaics (PV), wind and hydropower, and, in the buildings sector, by the installaƟon of heat pumps for space heaƟng. Demand for energy services is further reduced by higher prices resulƟng from subsidy removal, CO2 pricing and more expensive electricity generaƟng technologies in the power sector.

Trends by region Roughly half of the global eĸciency-related energy savings in the New Policies Scenario are achieved in China, North America and Europe. This reŇects not only their current and expected shares in total global energy demand, but also their emphasis on eĸciency (Tableථ7.2). In 2035, the largest share of the diīerenƟal energy savings between the Current and New Policies Scenarios is achieved in China, represenƟng about 40% of the global total (Figureථ7.6). Due to energy eĸciency, Chinese energy demand growth from 2011 to 2035 slows from an annual average of 2.2% to 1.9%. Energy eĸciency, together with structural changes in the economy, contributes to a decline in China’s energy intensity of 60% between 2011 and 2035. The main contribuƟng factor to energy savings in China in the New Policies Scenario, compared to the Current Policies Scenario, is the more intense shiŌ in the Chinese economy from energy-intensive industries to light industry and services. North America achieves the second-largest savings, at roughly 190ථMtoe in 2035, almost halving the annual growth in its energy demand to 2035 compared with the Current Policies Scenario. North American energy intensity declines by about 40%, based on more ambiƟous energy eĸciency policies in transport, industry and buildings. As a result, the gap between energy intensity in North America and OECD Europe narrows: in 2011 North America was using 49% more energy than Europe to produce one unit of GDP; by 2035 it uses 37% more. Europe’s energy demand in the New Policies Scenario is 7% lower than in the Current Policies Scenario (about 120ථMtoe in 2035). The main driver behind the energy savings in Europe is the implementaƟon of the EU Energy Eĸciency DirecƟve. The main components that reduce Įnal energy consumpƟon are the energy eĸciency obligaƟon scheme, together with, the renovaƟon of the building stock, accurate and individual billing, and mandatory energy audits in industry.

© OECD/IEA, 2013

In the New Policies Scenario, India’s total primary energy demand more than doubles over the Outlook period; annual growth averages 3.0%, compared with 3.3% in the Current Policies Scenario. A key driver of the reducƟon is the assumed extension of the current Perform, Achieve and Trade (PAT) scheme, which includes Įnancing mechanisms to support 10.ഩ Demand for energy services (e.g. for transport, lighting or space heating) is different from final energy consumption. The latter is a result of the fuel used and the efficiency of the end-use device chosen to satisfy the energy service demand. 242

World Energy Outlook 2013 | Global Energy Trends

the implementaƟon of eĸciency measures. Other developing countries in Asia, parƟcularly ASEAN countries, have recently stepped-up eīorts to improve energy eĸciency, using regulaƟons, market-based instruments and Įnancial incenƟves. Compared to the Current Policies Scenario, other developing countries in Asia save a total of around 60ථMtoe in the New Policies Scenario in 2035, with almost 60% of the savings stemming from ASEAN countries (IEA, 2013b). Persistent subsidies for fossil fuels in several countries in the region increase the payback periods of energy eĸciency measures and thus impede eĸciency gains.

toe per thousand dollars of GDP ($2012, MER)

Figure 7.6a ‫ ( ٲ‬QHUJ\LQWHQVLW\E\UHJLRQLQWKH1HZ3ROLFLHV6FHQDULR 0.5

2012 2035

0.4

1 2 3 4 5 6

0.3

7

0.2

8

0.1

9 OECD OECD OECD EEE AMS ASOC Europe

China India

ODA

Lan Africa Middle World America East

10

Figure 7.6b ‫ ٲ‬3  ULPDU\HQHUJ\VDYLQJVE\UHJLRQLQWKH1HZ3ROLFLHV6FHQDULR UHODWLYHWRWKH&XUUHQW3ROLFLHV6FHQDULRLQ OECD OECD OECD AMS ASOC Europe EEE

China India

12

Lan Middle ODA America Africa East

13

-50

14

-100 -150

15

© OECD/IEA, 2013

Mtoe

-200

16

-500 -550

Notes: OECD AMS с OECD Americas; OECD ASOC с OECD Asia Oceania; EEE с Eastern EuropeͬEurasia; ODA с Other Developing Asia.

Chapter 7 | Energy efficiency outlook

11

243

17 18

Table 7.2 ‫ ٲ‬.  H\HQHUJ\HIÀFLHQF\DVVXPSWLRQVLQPDMRUUHJLRQVLQWKH1HZ 3ROLFLHV6FHQDULRV New Policies Scenario United States

Fuel-economy standards for new passenger light-duty vehicles (PLDV) at 54.5 miles per gallon (4.3 litres per 100 kilometres ΀lͬkm΁) in 2025, and conƟnued improvement thereaŌer. Fuel-economy standards for new trucks (up to 20% by 2017ͬ2018, depending on type). Increased state and uƟlity budgets for energy eĸciency and conƟnued improvement thereaŌer. MEPS extended and strengthened for a number of appliances.

European Union

ParƟal implementaƟon of the Energy Eĸciency DirecƟve by 2020 and improvements thereaŌer. MEPS for buildings, as speciĮed in the Energy Performance for Buildings DirecƟve.* Mandatory environmental performance requirements for energy-using products, as speciĮed by the Ecodesign DirecƟve.** Fuel-economy standards for PLDVs at 95ථgථCO2ͬkm in 2020 (3.8 lͬ100 km), light-commercial vehicles at 147ථgථCO2ͬkm (5.9 lͬ100 km) in 2020.

Japan

Measures to contain electricity demand growth. Mandatory energy eĸciency benchmarking and energy management.* Eĸcient lighƟng, net zero energy use for all new buildings from 2030, Top-Runner Programme. Fuel-economy standards for PLDVs at 20.3 kmͬl (4.9 lͬ100 km) in 2020* and strengthening thereaŌer.

China

Target in the 12th Five-zear Plan (2011-2015) to cut energy intensity by 16% including: - ShiŌ towards a more service-oriented economy. - Top-10ථ000 Energy-consuming Enterprises Programme.** - Small plant closure and phasing out of outdated capacity.* - IncenƟves for buildings refurbishment, targets for energy eĸciency in buildings.** Fuel-economy standards for PLDVs at 5.0ථlͬ100ථkm in 2020, and strengthening thereaŌer. Fossil-fuel subsidies phased out within ten years.

India

Full implementaƟon and extension of the NaƟonal Mission on Enhanced Energy Eĸciency.* Fuel-economy standards for PLDVs: assumed annual improvement of 1.3% over 2012-2020. ImplementaƟon of MEPS and labelling for equipment and appliances, as well as support of eĸcient lighƟng.* Fossil-fuel subsidies phased out within ten years.

Brazil

Phasing out of incandescent lighƟng.** Measures to increase eĸciency in appliances and air condiƟoners.** Enhanced acƟon according to the naƟonal energy eĸciency plan. Inovar-Auto iniƟaƟve targeƟng fuel eĸciency improvement for PLDVs of at least 12% in 2017.**

© OECD/IEA, 2013

Middle East

ParƟal phase out of fossil fuel subsidies. Measures to increase eĸciency of lighƟng, appliances and air condiƟoners.

*ථAlso implemented in the Current Policies Scenario. **ථParƟally implemented in the Current Policies Scenario. Note: All fuel-economy standards refer to test-cycle fuel consumpƟon.

244

World Energy Outlook 2013 | Global Energy Trends

Trends by sector Global primary energy demand in the New Policies Scenario in 2035 is 1ථ260ථMtoe, or 7%, lower than in the Current Policies Scenario. Slightly more than half of the primary energy savings come from the power sector. Only a small porƟon of the savings in the power sector comes from improved eĸciency in the sector itself, the vast majority of the savings being aƩributable to lower electricity consumpƟon in buildings and industry (Figureථ7.7). Once the electricity savings are recalculated into primary energy terms (accounƟng for conversion losses), industry saves the most (37%), followed by transport (31%) and buildings (26%). While electricity savings dominate the industry and buildings sectors, oil is the dominant fuel in transport and shows the highest savings. Figure 7.7 ‫ ٲ‬3  ULPDU\HQHUJ\VDYLQJVIURPHQHUJ\HIÀFLHQF\E\IXHODQG

Coal

5

9

Oil Bioenergy

10

Other renewables

11

Transport 50

100

150

200

250

300

12

350 Mtoe

*ථElectricity and heat demand savings in end-use sectors are converted into equivalent primary energy savings and aƩributed to each end-use. The savings allocated to the power sector arise from the increased eĸciency of the plant and of grid and system management (eĸciency improvements in transmission and distribuƟon are labelled as ͞electricity and heat͟ in power). Note: Energy savings of 17ථMtoe in agriculture are not depicted, while non-energy use does not have any eĸciency-related savings between the two scenarios.

The industry sector is responsible for 30% of global Įnal energy consumpƟon and one- third of energy-related CO2 emissions (including indirect emissions from electricity and heat). Over the projecƟon period, industrial energy consumpƟon grows to some 3ථ530ථMtoe from today’s level of 2ථ550ථMtoe (Tableථ7.3). Most of this growth arises in non-energyintensive sectors, such as food, texƟles, machinery and transport equipment, whose energy consumpƟon increases by 60%, while energy consumpƟon in iron and steel grows by only 16% and is almost Ňat in cement. Chapter 7 | Energy efficiency outlook

13 14 15

Industry

© OECD/IEA, 2013

4

8

Gas

Buildings

3

7

Electricity and heat*

Industry

2

6

VHFWRULQWKH1HZ3ROLFLHV6FHQDULRUHODWLYHWRWKH&XUUHQW 3ROLFLHV6FHQDULRLQ

Power

1



16 17 18

Table 7.3 ‫ ٲ‬6 DYLQJVLQLQGXVWULDOHQHUJ\GHPDQGDQG&22HPLVVLRQVIURP HQHUJ\HIÀFLHQF\LQWKH1HZ3ROLFLHVVFHQDULR(Mtoe) Change versus Current Policies Scenario Demand

Total

Due to eĸciency

2011

2020

2035

2020

2035

2020

2035

Coal

727

848

814

-27

-73

-11

-35

Oil

324

355

348

-10

-30

-7

-20

Gas

502

624

783

-14

-61

-12

-46

Electricity

671

885

1134

-27

-98

-20

-66

Heat

126

143

148

-2

-10

-3

-9

Bioenergy*

199

242

300

-2

6

-6

-18

2ථ548

3ථ096

3ථ528

-82

-265

-59

-194

9.7

11.8

12.3

-0.5

-2.0

-0.3

-0.9

Total CO2 emissions (Gt)**

* Includes other renewables. ** CO2 emissions include indirect emissions from electricity and heat; Gt с gigatonnes.

Compared with the Current Policies Scenario, industrial energy consumpƟon is 7% lower in the New Policies Scenario in 2035. Almost three-quarters of the reducƟon is driven by energy eĸciency gains. Most of the new policies currently under discussion focus on energy audits, energy management systems and Įnancial incenƟves, parƟcularly for small and medium-size enterprises. Many of these companies are in non-energy-intensive industries, as heavy industries are generally dominated by large corporaƟons. Consequently, three-quarters of the eĸciency-related savings come from light industries in 2035. The opƟmisaƟon of electric motor systems, including the increased adopƟon of variable speed drives, is responsible for most of the electricity savings, given the large share of these systems in consumpƟon and their improvement potenƟal (IEA, 2013c).

© OECD/IEA, 2013

Reduced demand for energy services accounts for about one-quarter of the energy savings in 2035 in the New Policies Scenario, compared with the Current Policies Scenario. Most of these demand-related savings occur in China as a result of a shiŌ from heavy industries to lighter ones and to the services sector. While demand declines in regions that phase out fossil-fuel subsidies, it slightly increases elsewhere, due to lower energy prices. The eīect of fuel and technology switching is minimal, as the posiƟve eīect from technology switching (e.g. from primary to secondary steel making) is almost completely oīset through the growth in the share of bioenergy, which is usually less eĸcient than fossil fuels. The steel sector11 is the world’s second most energy-consuming industry (aŌer chemicals and petrochemicals), consuming as much energy each year as Russia. There is over-capacity in steel making, and since energy is an important cost factor, eĸcient energy use is a key determinant of plant compeƟƟveness (see Chapterථ8). Energy intensity can be reduced in many ways: by adopƟng more eĸcient technologies, by opƟmising and managing systems, 11.ഩ For the purpose of this analysis, the steel sector is defined to include coke ovens and blast furnaces. 246

World Energy Outlook 2013 | Global Energy Trends

and by process change (IEA, 2012a). In the steel sector, process changes include a shiŌ towards the use of scrap metal or direct reduced iron instead of pig iron for iron making, and a shiŌ from primary steel making via blast furnaces (BF) and basic oxygen furnaces (BOF) (or open hearth furnaces) to the use of electric arc furnaces (EAFs). Process change is parƟcularly important in countries where a signiĮcant share of steel is produced via the BFͬBOF route and where scrap metal is available. These include Japan, the European Union, Russia and China. Steel producƟon from EAFs using scrap metal is less than half as energy intensive as the BFͬBOF route and most steel products nowadays can be produced in EAFs. While energy costs and environmental consideraƟons inŇuence the choice of whether to use EAFs, scrap metal availability and steel quality are stronger factors. In the New Policies Scenario, process changes in the steel industry play an important role in reducing energy intensity up to 2035, accounƟng for slightly more than half of all energy savings (Figureථ7.8). This means that improved energy eĸciency is responsible for less than half of the energy intensity reducƟon. Figure 7.8 ‫ ( ٲ‬QHUJ\LQWHQVLW\UHGXFWLRQLQWKHLURQDQGVWHHOVHFWRUE\W\SHRI LPSURYHPHQW United States

Japan

European Union

Russia

1 2 3 4 5 6 7 8

China

India

9

Technical efficiency -5%

Systems opmisaon

10

Process change -10%

11

-15%

12

-20% -25%

13

-30%

14

© OECD/IEA, 2013

Note: These regions account for almost 80% of global steel producƟon.

Russia achieves signiĮcant energy savings by moving away from outdated open hearth furnace technology. Steel producƟon in China is currently dominated by BOFs. Increasing the use of scrap metal in EAFs, as well as raising the share of steel producƟon from EAFs by nine percentage points reduces the energy consumpƟon per unit of steel by 9%. For Japan, we project in the New Policies Scenario a slightly higher domesƟc consumpƟon of scrap metal and higher producƟon share from EAFs to a similar level as in the late 1990s. In the European Union, increasing the share of secondary steel making from the current 43% to 58% in 2035 and higher use of scrap metal are responsible for the majority of the energy intensity reducƟon. The United States and India do not make a big contribuƟon to energy Chapter 7 | Energy efficiency outlook

247

15 16 17 18

savings through process change because steel making via the EAF route already accounts for around 60% in both countries. In India, most of the input to EAFs is from coal-based direct reduced iron, which is signiĮcantly less eĸcient than the gas-based variant. The remaining energy intensity reducƟons in steel producƟon come from the adopƟon of more eĸcient technical equipment and systems opƟmisaƟon via process control, automaƟon and monitoring systems. Systems improvements account for roughly oneĮŌh of the non-process related eĸciency gains. The intensity reducƟons from technical eĸciency are parƟcular signiĮcant in Russia, India and the United States. Further eĸciency improvements in China are limited, mainly due to the limited amount of new capacity addiƟons. China added substanƟal steel capacity over recent years and domesƟc steel producƟon is expected to peak around 2020, with declining producƟon thereaŌer.

Transport Almost 30% of global Įnal energy consumpƟon is in the transport sector. This sector is heavily reliant on oil, with the notable excepƟons of electricity in rail networks and natural gas in the operaƟon of pipelines. In the New Policies Scenario, energy consumpƟon increases by 1.3% annually to reach 3ථ300ථMtoe in 2035, which compares to a growth rate of 2.1% per year over the past twenty years. The slowdown in growth mainly reŇects trends in road transport, but growth in rail travel and aviaƟon outpaces growth in road travel in the coming decades. Transport-related CO2 emissions increase from 7ථgigatonnes (Gt) in 2011 to 9ථGt in 2035, which is the fastest increase of all end-use sectors (Tableථ7.4). Table 7.4 ‫ ٲ‬6 DYLQJVLQWUDQVSRUWHQHUJ\GHPDQGDQG&22HPLVVLRQVIURP HQHUJ\HIÀFLHQF\LQWKH1HZ3ROLFLHV6FHQDULR (Mtoe) Change versus Current Policies Scenario Demand 2011 Coal

Total 2035

2020

Due to eĸciency 2035

2020

2035

3

2

0

0

0

0

0

2ථ264

2ථ572

2ථ878

-47

-348

-33

-248

Gas

93

122

186

5

39

-2

-10

Electricity

25

35

63

1

12

-1

-1

Biofuels

59

101

192

13

40

-1

-19

2ථ444

2ථ832

3ථ319

-27

-258

-38

-278

7.0

8.0

9.0

-0.1

-0.9

-0.1

-0.8

Oil

Total CO2 emissions (Gt)

© OECD/IEA, 2013

2020

Transport energy demand in the New Policies Scenario is about 260ථMtoe, or 7%, lower in 2035 than in the Current Policies Scenario. Higher eĸciency in transport decreases consumpƟon by about 280ථMtoe in 2035, mainly as a result of stricter fuel-economy standards for light-duty vehicles. In several regions in the New Policies Scenario – namely Southeast Asia, LaƟn America, the Middle East and China – eĸciency improvements are partly oīset by the rebound eīect, increased demand for road travel as a result of lower oil prices. 248

World Energy Outlook 2013 | Global Energy Trends

The approach most commonly used to improve energy eĸciency in road transport is the introducƟon of fuel-economy standards. The impact of such policies considered in the New Policies Scenario on total fuel savings in road transport, are greatest in China, followed by the United States and the European Union (Figureථ7.9). In China, under the State Council’s Development Plan for Energy Saving and the Automobile Industry for 2012-2020, the aim is to reach an average level of fuel consumpƟon of 5.0ථlitres per 100 kilometres (lͬ100km) for new cars, saving roughly 12ථMtoe in 2020 in the New Policies Scenario relaƟve to the Current Policies Scenario. In the United States, the Corporate Average Fuel Eĸciency (CAFE) standards applicable to 2025 save 61ථMtoe by 2035, relaƟve to the Current Policies Scenario. These standards may undergo mid-term evaluaƟon, thus making their outlook uncertain not only in the United States but also in Canada, which has recently adopted similar standards. In the European Union, the target of reaching an average emissions level of 95 grammes of CO2 per kilometre (gථCO2ͬkm) in 2020 for passenger light-duty vehicles (PLDVs), which is sƟll awaiƟng approval from the European Council, is responsible for the majority of savings in the New Policies Scenario, totalling 24ථMtoe in 2035.

1 2 3 4 5 6 7

Figure 7.9 ‫ ) ٲ‬XHOVDYLQJVIURPHQHUJ\HIÀFLHQF\LQURDGWUDQVSRUWLQWKH1HZ

8

3ROLFLHV6FHQDULRUHODWLYHWRWKH&XUUHQW3ROLFLHV6FHQDULR 2010

2015

2020

2025

2030

2035

9

United States Canada -50

European Union

10

China -100

India

11

Brazil -150

Other OECD Other non-OECD

© OECD/IEA, 2013

Mtoe

-200

13

-250

Compared with road transport, the energy eĸciency of other transport sectors has so far come under less scruƟny, but recently some encouraging developments have taken place. AŌer aviaƟon was included in the EU Emissions Trading Scheme in 2012, internaƟonal airlines announced, for the Įrst Ɵme, in early 2013 their readiness to curb their greenhouse-gas emissions. The InternaƟonal Air Transport AssociaƟon issued a resoluƟon urging governments to manage CO2 from air travel from 2020 onwards, using a marketbased mechanism. The goal of the associaƟon is to increase fuel eĸciency on average by 1.5% per year unƟl 2020, broadly in line with the eĸciency improvements achieved in the New Policies Scenario (IATA, 2009). In addiƟon, the United NaƟons InternaƟonal Civil AviaƟon OrganisaƟon also reached consensus in October 2013 on a roadmap to create a market-based mechanism to reduce carbon emissions. In the mariƟme sector, the major iniƟaƟve to improve fuel eĸciency is the Energy Eĸciency Design Index for new ships. Chapter 7 | Energy efficiency outlook

12



14 15 16 17 18

Adopted by the InternaƟonal MariƟme OrganizaƟon in 2011, it foresees the applicaƟon of progressively more stringent eĸciency targets. The index entered into force in January 2013 and applies to all ships of 400 gross tonnage and above (which account for 70% of CO2 emissions from ships), though exempƟon from the requirements for new ships may be available for up to a maximum of four years (Hughes, 2013).

Buildings Final energy demand in the buildings sector grows from 2ථ890ථMtoe in 2011 to 3ථ690ථMtoe in 2035, 180ථMtoe or 5% less than in the Current Policies Scenario (Tableථ7.5). Most of the Įnal energy savings arise from energy eĸciency measures targeƟng the building shell, but also from eĸciency standards for lighƟng and other energy-consuming equipment, such as heaƟng systems and appliances, as well as from beƩer use of automaƟon and control systems. A lower call for energy services is parƟcularly important in countries phasing out fossil-fuel subsidies for households, including China and Russia, with fuel and technology switching playing a smaller role. Table 7.5 ‫ ٲ‬6 DYLQJVLQEXLOGLQJVHQHUJ\GHPDQGDQG&22HPLVVLRQVIURP HQHUJ\HIÀFLHQF\LQWKH1HZ3ROLFLHVVFHQDULR(Mtoe) Change versus Current Policies Scenario Demand

Total

Due to eĸciency

2011

2020

2035

2020

2035

2020

2035

Coal

118

117

95

-5

-16

-1

-2

Oil

324

318

271

-11

-33

-3

-8

Gas

597

689

815

-12

-44

-5

-21

Electricity

839

1ථ044

1ථ417

-29

-105

-13

-64

Heat

149

158

167

-3

-9

-1

-4

Modern renewables*

115

157

243

6

38

0

-1

TradiƟonal biomass**

744

730

680

-2

-9

-1

-4

2ථ886

3ථ213

3ථ688

-57

-178

-26

-103

8.1

8.7

9.4

-0.5

-1.9

-0.2

-0.7

Total CO2 emissions (Gt)***

© OECD/IEA, 2013

*ථModern renewables include wind, solar and geothermal energy as well as modern, eĸcient biomass use. **ථTradiƟonal biomass includes fuelwood, charcoal, animal dung and some agricultural residues. ***ථCO2ථemissions include indirect emissions from electricity generaƟon and energy use for heat.

Around 60% of the savings between the Current Policies and New Policies Scenarios in 2035 are in electricity, with the amount saved represenƟng more than the current annual generaƟon in Japan. Despite the large absolute savings, electricity becomes a more important energy carrier over Ɵme, increasing its share of buildings energy use from 29% today to 38% in 2035, as the relaƟve importance of direct use of fossil fuels and tradiƟonal biomass declines. Space and water heaƟng contribute most to these savings in the New Policies Scenario, closely followed by appliances and cooling. Savings in coal and oil use account for around 30% of total savings, mainly driven by beƩer insulaƟon, which increases the eĸciency of buildings (and thus reduces demand for space heaƟngͬcooling) and the 

World Energy Outlook 2013 | Global Energy Trends

uptake of more eĸcient water heaters and cookstoves in developing countries. TradiƟonal biomass sƟll makes up 26% of Įnal energy consumpƟon in buildings today, mostly used for cooking in developing countries in fairly simple and mostly ineĸcient cookstoves (see Chapter 2). With increasing household income and poliƟcal support, modern fuels including charcoal, kerosene, liqueĮed petroleum gas (LPG), gas and electricity could gradually take the place of tradiƟonal biomass (IEA, 2010). In the services sector, the energy consumpƟon per unit of value added has been declining in all regions and this trend is projected to conƟnue, reaching 70% of its current value in 2035. In the residenƟal sector, energy intensity, deĮned as consumpƟon of modern fuels12 per square metre and per capita, is projected to decrease by 39% in total, but shows larger regional diīerences. In OECD countries, energy intensity in residenƟal buildings decreases by 25%, mainly driven by eĸciency improvements. In non-OECD countries, energy intensity decreases by more than 30%, while residenƟal Ňoor space increases by 60% by 2035. In emerging economies, such as China, India and ASEAN countries, eĸciency improvements are partly oīset by a signiĮcantly increased demand for energy services, reŇecƟng increased income per capita. In poorer developing countries, improved access to electricity and reduced use of tradiƟonal biomass for cooking also play important roles: increasing electricity access leads to increased modern fuel consumpƟon, but the low consumpƟon of newly connected households reduces the energy intensity of the residenƟal sector at a naƟonal level.

© OECD/IEA, 2013

China achieves about one-third of the global residenƟal energy savings diīerenƟal between the Current Policies Scenario and New Policies Scenario, thanks to policies under discussion and energy price reforms (Spotlight). Whether countries realise their economic potenƟal for savings is closely linked to the structure of the buildings sector and the condiƟons governing energy supply. Savings in space heaƟng in the residenƟal sector in China, for example, remain below the full economic potenƟal, due to the Ňat rate tariī structure of district heaƟng systems in northern urban areas, which is a barrier to eĸciency changes (Figureථ7.10). More stringent standards in appliances and air condiƟoners are expected to deliver most of the savings in the residenƟal sector. Cooling needs in China and other developing countries in warm climate zones are expected to increase strongly over Ɵme as people become more aŋuent, highlighƟng the importance of the prompt introducƟon of relevant eĸciency standards. Cooling needs also rise due to climate change (IEA, 2013c). In Europe, the savings between the Current and New Policies Scenario account for more than 15% of the global savings in households in 2035, as policies under discussion take eīect. InsulaƟon and retroĮt measures provide most of the gains, as a consequence of stricter renovaƟon policies and building codes under the EUථEĸciency DirecƟve (the impact of the Energy Performance of Buildings DirecƟve is mainly integrated into the Current Policies Scenario). The implementaƟon of building codes is mandatory in most OECD countries, but, while progress has been made in non-OECD countries, implementaƟon usually remains

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18

12.ഩ Excluding traditional biomass.

Chapter 7 | Energy efficiency outlook



voluntary there. Achieving high compliance or designing eīecƟve building codes can be very challenging. Policies currently in place tend to encourage acƟon by individuals, such as the replacement of windows or boilers, which has only a limited impact on the overall energy consumpƟon, compared to holisƟc measures, such as deep renovaƟon of the building shell (Saheb, 2013). Consequently, the impact of current policies on building’s energy consumpƟon remains limited, as large potenƟal savings remain untapped. Nevertheless, recent analysis shows that the market is starƟng to reward increased building energy standards: they have been shown to contribute to higher rents and house prices in Europe (BIO, IEEP and Lyons, 2013). The second-largest savings in the European Union come from improvements in appliances and lighƟng, supported by the EU Ecodesign DirecƟve. In the United States, increased eĸciency standards for appliances and air condiƟoners are expected to deliver most of the savings in the residenƟal sector, together with savings in space and water heaƟng, thanks to building codes.

S P O T L I G H T 3ROLFLHVIRUHQHUJ\HIILFLHQF\LQEXLOGLQJVLQ&KLQD China accounts for about one-third of the global savings in the buildings sector between the Current and the New Policies Scenario. This is primarily due to the policies announced in the 12th Five-zear Plan and the central co-ordinaƟon of building regulaƟons in the Ministry of Housing and Rural Urban Development (MOHURD).

© OECD/IEA, 2013

China’s 12th Five-zear Plan introduced the concept of regulaƟng the absolute level of energy demand, rather than energy intensity. Although no cap has yet been announced, this can be seen as an important shiŌ in Chinese energy policy. In May 2012, the MOHURD announced the Building ConservaƟon Plan, which aims to contribute to the overall naƟonal intensity target for 2015 by means of eĸciency savings in the buildings sector. The plan provides that 95% of new buildings should reduce space heaƟng per square metre (mϸ) by 55-65%, compared with 1980 levels, depending on the climate zone. Furthermore, in the period to 2015, 450ථmillionථsquare metres of exisƟng buildings are to be refurbished (1% of the exisƟng stock) and 800ථmillionථmϸ of new ͞green buildings͟ built. The criteria for green buildings include resource savings and environmental protecƟon; examples are the deployment of solar PV or natural lighƟng (Tsinghua University, 2013). The MOHURD regulates the largest building industry in the world: growth in Chinese residenƟal Ňoor space accounted for more than one-third of the Ňoor space addiƟons worldwide in the last ten years. The MOHURD reports to the state council and supervises provincial building departments. This structure allows the regulator to follow regional developments closely and ensures high compliance with building regulaƟons (up to 95% and higher has been observed) (hou, McNeil and Levine, 2011).



World Energy Outlook 2013 | Global Energy Trends

Figure 7.10a ‫ ( ٲ‬QHUJ\FRQVXPSWLRQE\HQGXVHLQKRXVHKROGVE\UHJLRQLQ WKH1HZ3ROLFLHV6FHQDULRDQG

100%

European Union

United States

China

India

Africa

Russia

1 2

World

3

80% 60%

4

40% 20%

5 2011 2035 2011 2035 2011 2035 2011 2035 2011 2035 2011 2035 Space and water heang

Cooling

Modern cooking

Tradional biomass*

2011 2035

6

Lighng and appliances

Figure 7.10b ‫) ٲ‬LQDOHQHUJ\VDYLQJVLQKRXVHKROGVLQWKH1HZ3ROLFLHV6FHQDULR

Mtoe

UHODWLYHWRWKH&XUUHQW3ROLFLHV6FHQDULRE\UHJLRQ 10

Lighng and appliances

0

Tradional biomass*

-10

8 9

Modern cooking Cooling

-20

7

Space and water heang

10 11

-30

12

-40 -50

13

-60 European Union

United States

China

India

Africa

Russia

14

© OECD/IEA, 2013

* Due to a lack of comprehensive data on the end-use of tradiƟonal biomass, it is assumed to be used primarily for cooking, although it also serves space and water heaƟng purposes.

In developing countries, e.g. in Africa and India, the increase in demand for electricity resulƟng from greater access to electricity oīsets the savings due to limited eĸciency improvements in lighƟng and appliances. Important savings in lighƟng can be achieved by phasing out the use of incandescent bulbs. LegislaƟon to this eīect is in place in all major OECD countries and in a number of non-OECD countries, including China, India, most countries of West Africa and Brazil. This single change leads to a reducƟon in electricity demand for lighƟng of around 5% in these regions, between the Current and New Policies Scenarios. If all increases in lighƟng demand were met by the most eĸcient technology, Chapter 7 | Energy efficiency outlook



15 16 17 18

global electricity demand from lighƟng could be reduced by more than 40% in 2035, resulƟng in a reducƟon of 5% of total residenƟal electricity demand. In countries with limited access to electricity, lower demand as a result of improved eĸciency by exisƟng customers provides an important oīset to addiƟonal electricity generaƟon.

Investment in energy eĸciency The Current Policies Scenario requires investment of $4.7ථtrillion in energy eĸciency over 2013-2035 (Figureථ7.11).13 Despite a diminishing share in global energy consumpƟon, OECD countries account for 60% of these total investments, due to more stringent policies regulaƟng eĸciency in the OECD. In order to realise the energy savings in the New Policies Scenario, addiƟonal investment of $3.4ථtrillion is needed in energy eĸciency. As a result of lower electricity demand, cumulaƟve investment in electricity transmission and distribuƟon is $0.8ථtrillion lower in the New Policies Scenario than in the Current Policies Scenario, and investment in power plants is reduced by $0.6ථtrillion, more than compensaƟng for the addiƟonal investment required to improve energy eĸciency in the New Policies Scenario. The $3.4ථtrillion addiƟonal investments in the New Policies Scenario generate savings on energy expenditures of $6.1ථtrillion up to 2035. Figure 7.11 ‫ ٲ‬$  YHUDJHDQQXDOHQHUJ\HIÀFLHQF\LQYHVWPHQWE\VFHQDULR

Billion dollars (2012)

DQGVHFWRU 150

Addional energy efficiency investment in New Policies Scenario over Current Policies Scenario

120

Energy efficiency investment in Current Policies Scenario

90 60 30

Transport Buildings Industry OECD

Transport Buildings Industry Non-OECD

© OECD/IEA, 2013

The transport sector accounts for almost 40% of the cumulaƟve addiƟonal investments in the New Policies Scenario. This reŇects the large increase in the vehicle Ňeet (which almost doubles to 2.9ථbillion vehicles in 2035) and the average amount spent on energy eĸciency for each vehicle ($300) throughout the projecƟon period in the New Policies Scenario. Eĸciency investments in commercial and residenƟal buildings increase substanƟally, driven by higher insulaƟon standards for new buildings, a more widespread adopƟon of 13.ഩ Energy efficiency investment (excluding international bunkers) is used to denote expenditure on a physical good or service which leads to future energy savings, compared with the energy demand expected otherwise. 

World Energy Outlook 2013 | Global Energy Trends

eĸcient heaƟng technologies, and more eĸcient lighƟng, appliances and cooling systems. Investment in industrial energy eĸciency in 2035 increases by about 70%, or $26ථbillion, compared with the Current Policies Scenario, realising energy savings of almost 200ථMtoe in the same year. About half of the investment goes to steam systems and furnaces, while the other half is used to improve electric motor systems, appliances and lighƟng. Though various market imperfecƟons inhibit the adopƟon of energy eĸciency measures, energy prices play a key role in sƟmulaƟng the adopƟon of eĸcient technologies. With high energy prices in Europe and Japan, and persistent price subsidies in a few energy exporƟng countries (see Chapterථ2), the payback period for a more eĸcient technology can be up to nine Ɵmes higher depending on the region (Figureථ7.12). This means that, in some cases, the payback period can exceed the lifeƟme of the technology, making it uneconomic to invest in such a measure (even if transacƟon costs are not counted). Including transacƟon costs would double the payback period for certain countries and sectors (IEA, 2012a).

Years

Figure 7.12 ‫ ٲ‬3  D\EDFNSHULRGVIRUVHOHFWHGWHFKQRORJLHVDQGUHJLRQV 18

VSD

2 3 4 5 6 7 8

LED

15

1

Hybrid car

9

12 9

10

6

11

3 United States

Japan

European Union

Russia

China

India

12

Middle East

Notes: VSD с installaƟon of a variable speed drive in a compressed air system. LED с light-emiƫng diode replacing an incandescent light bulb. Hybrid car с hybrid car replacing an internal-combusƟon engine car.

© OECD/IEA, 2013

The economic incenƟve to improve eĸciency is highest in Japan and the European Union. Installing a variable speed drive (VSD), a device to control the speed of machinery, is one of the best ways of achieving energy savings in industrial electric motor systems, which account for up to 70% of all electricity consumpƟon in industry. As industrial electricity prices are highest in Japan and Europe, the payback period of a VSD is around one year, while it is around two-and-a-half years in the United States, where electricity prices are less than half of the European average. The payback period of a VSD is relaƟvely low in India, because the total cost of installing a VSD is lower, since labour costs are a fracƟon of OECD levels and account for around one-third of the total costs (UNIDO, 2010). LighƟng accounts for roughly 20% of the current electricity demand in buildings. Lightemiƫng diodes use around Įve Ɵmes less energy than incandescent light bulbs. Given the relaƟvely low electricity prices for households in India, China and Russia, the payback Chapter 7 | Energy efficiency outlook



13 14 15 16 17 18

period for installing LEDs, instead of incandescent light bulbs, is more than two years. Because of even lower electricity prices in the Middle East, the payback period there is around nine years. Oil prices are also low in the Middle East, meaning that payback for an investment in a hybrid car stretches to eighteen years, while in Russia where the annual mileage is relaƟvely low, it is nine years. Although gasoline prices are signiĮcantly lower in the United States than in Europe, higher vehicle mileage in the United States reduces the diīerence in payback periods in this case.

Broader beneĮts Improvements in energy eĸciency not only reduce energy consumpƟon and energy bills, but have further beneĮts including lower energy imports, macroeconomic advantages, and reduced levels of air polluƟon and CO2 emissions.14

Energy imports In net-imporƟng regions, improved energy eĸciency in the New Policies Scenario enhances energy security by reducing energy demand and thereby lowering energy import bills. In 2035, avoided import bills stemming from energy eĸciency in the New Policies Scenario are highest in China, at $130 billion, and the United States, at $95ථbillion. On a per-capita basis, however, avoided import bills are by far the highest in the United States, at $250ථper person in 2035 (Figureථ7.13). Most savings result from higher eĸciency in PLDVs, which reduces oil use. Natural gas-related savings are signiĮcant in some countries, such as Japan and
United States

Gas

Japan

Coal

Korea European Union China India

© OECD/IEA, 2013

Southeast Asia 40

80

120

160

200

240 280 Dollars per capita

14.ഩ For a comprehensive overview of the multiple benefits of energy efficiency, see IEA (2012b). 

World Energy Outlook 2013 | Global Energy Trends

Impact on total household expenditure The addiƟonal energy eĸciency measures implemented in the New Policies Scenario aīect the broader economy and alter consumpƟon paƩerns of all types of goods and services (Figureථ7.14). Firms increase investment in more energy-eĸcient producƟon processes, see a net reducƟon in their energy costs and spend more on non-energy inputs (e.g. capital and labour). These economy-wide changes in consumpƟon paƩerns follow adjustments in the relaƟve prices of diīerent commodiƟes. The prices of manufactured goods and services, which have less energy embedded, are reduced, sƟmulaƟng demand from Įrms and households.

2 3 4 5

Figure 7.14 ‫ ( ٲ‬FRQRPLFLPSDFWVRIHQHUJ\HIÀFLHQF\

6

Firms and household behaviour

7

Firms: energy efficiency investments Household: spending on energy efficient appliances and buildings insulation

Reform of fossilfuel consumption subsidies

1

8 9

Energy savings

10 Lower energy spending

Reduce energy imports

11 12

Firms: increased use of non-energy goods for production

Household: higher disposable income for nonenergy purchases

Trade balance improvements

13 14

Increased demand for non-energy goods by firms and households

© OECD/IEA, 2013

Increased activity and employment in non-energy sectors

15

Feedback (rebound effects)

16

Higher activity in non-energy sectors

Households: increase in real disposable income

Chapter 7 | Energy efficiency outlook

17 18

GDP impact



In parallel, households purchase more eĸcient electrical appliances, buy more eĸcient cars and refurbish their homes to improve insulaƟon. SƟmulated acƟvity in non-energy manufacturing creates higher labour demand. This shiŌ in employment beneĮts workers through higher wages and therefore translates into higher disposable income. The overall macroeconomic impact of improving energy eĸciency is generally posiƟve, as illustrated in the Eĸcient World Scenario (IEA, 2012a). In value terms, household expenditure currently accounts for more than half of global GDP. In OECD countries, where 35% of total Įnal energy savings are achieved in the New Policies Scenario, the share of GDP household consumpƟon is close to 70%. Moreover, it is in residenƟal energy demand and in PLDV fuel demand, the vast bulk of which is consumed by households, that 37% of total eĸciency savings are realised. Energy eĸciency measures adopted in the New Policies Scenario result in a net posiƟve outcome for households: the cumulaƟve reducƟon in household energy spending through to 2035 reaches $2.6ථtrillion (Figureථ7.15), corresponding to about 3% of current global GDP. Reduced energy spending is accompanied by reduced fuel prices in the New Policies Scenario, which in turn bring down other producƟon costs. These energy savings free up disposable income, some of which is allocated to the consumpƟon of cheaper nonenergy goods, including energy-eĸcient appliances. Almost 60% of the reducƟon in energy spending is made by OECD households, while China accounts for another one-quarter. However, only a small share of addiƟonal purchases of non-energy goods occur in OECD countries, mainly because the same amount of disposable income for households in less developed countries, with lower iniƟal endowments, induces larger shiŌs in consumpƟon paƩerns. Furthermore, a degree of saturaƟon in the consumpƟon of certain durable goods in OECD countries leaves less room for altering OECD consumpƟon paƩerns.

© OECD/IEA, 2013

European households currently pay among the highest energy prices in the world and their total average spending on energy is more than 10% of their total spending (see Chapter 8), making energy eĸciency a parƟcularly aƩracƟve opƟon. In the New Policies Scenario, every dollar saved per person in the European Union on energy bills in 2020 is largely oīset by addiƟonal spending on other goods. Over the projecƟon period, European households see their purchasing power essenƟally unchanged. Japanese households who spend a lower share of their income on energy (about 8%) make modest gains, in the order of $30ථbillion. Chinese households see their net real consumpƟon increasing by about $100ථbillion over the period. This amount, equivalent to just 0.2% of China’s GDP in 2035, is insuĸcient to trigger a sizeable shiŌ in private consumpƟon in China. In other fast-growing non-OECD countries, such as India and Indonesia, households with the lowest incomes can rarely aīord to upgrade ineĸcient home installaƟons. Hence the potenƟal for direct energy savings in the residenƟal sector is lower. AddiƟonally, the impact of eĸciency policies on energy spending is oŌen limited, as in the New Policies Scenario these measures are assumed to be complemented by reforms to fossil-fuel subsidies (Chapterථ2). The combined eīect of the two sets of measures leaves overall household expenditures largely unchanged. 

World Energy Outlook 2013 | Global Energy Trends

Figure 7.15 ‫& ٲ‬  KDQJHLQDQQXDOSHUFDSLWDKRXVHKROGVSHQGLQJRQHQHUJ\ DQGQRQHQHUJ\JRRGVDQGVHUYLFHVLQWKH1HZ3ROLFLHV 6FHQDULRUHODWLYHWRWKH&XUUHQW3ROLFLHV6FHQDULR

Dollar per capita

World

Japan

European Union

2

China

150

Non-energy Energy

100

1

3 4

50 0

5

-50

6

-100

2035

2020

2012

2035

2020

2012

2035

2020

2035 2012

2020

7 2012

-150

8

Sources: IEA analysis using the World Energy Model and OECD analysis using the OECD ENV-Linkages model.

9

>ŽĐĂůĂŝƌƉŽůůƵƟŽŶ As a result of lower demand for fossil fuels, improved energy eĸciency leads to reduced local air polluƟon, which helps to reduce respiratory diseases. Recent analysis shows that each year more than three million people die from outdoor air polluƟon, mainly caused by combusƟon of fossil fuels and bioenergy (Lim, et al., 2012). Today, China and India are responsible for more than 40% of global sulphur dioxide (SO2) emissions, of which more than one-quarter arise from coal power plants. SO2 emissions are reduced by 10% in the New Policies Scenario as a result of eĸciency improvements with more than 70% of the reducƟon achieved in the power sector of non-OECD countries. The largest source of nitrogen oxides (NOX) emissions is currently the transport sector (around 50%), followed by power and industry. The transport sector accounts for 20% of the reducƟon in NOX emissions, with the bulk of the improvements seen in non-OECD countries. ParƟculate maƩer (PM2.5) emissions are caused by the use of tradiƟonal biomass and industrial processes. PM2.5 emissions are reduced by 4% globally resulƟng from a reducƟon of tradiƟonal biomass use as more clean cooking faciliƟes are adopted.15

© OECD/IEA, 2013

11 12 13 14 15 16

CO2 emissions In the New Policies Scenario energy-related CO2 emissions increase from 31.5ථGtථ in 2012 to 37.2ථGt in 2035, which is about 5.9ථGt or 14% lower than in the Current Policies Scenario (Figureථ7.16). Energy eĸciency measures account for about half of cumulaƟve 15.ഩ This is estimated based on IIASA (2012).

Chapter 7 | Energy efficiency outlook

10



17 18

CO2 emissions savings in the New Policies Scenario, with the share being even higher in the short term. Energy eĸciency measures, including in lighƟng and electric motor systems, are in most cases quickly deployable and among the cheapest opƟons to reduce CO2 emissions. Redrawing the Energy-Climate Map, a WEO special report, proposes a set of four pragmaƟc policies that can have a signiĮcant impact by 2020, with no net cost to the overall economy (see Chapterථ2). Energy eĸciency on its own is, however, not enough to bring down emissions to the level compaƟble with limiƟng the long-term temperature increase to 2ථΣC. It needs to be complemented by increased use of renewables across all sectors and wider deployment of nuclear power, carbon capture and storage (CCS) in power generaƟon and industry, and electric vehicles in transport. Figure 7.16 ‫ ٲ‬:  RUOGHQHUJ\UHODWHG&22HPLVVLRQVDEDWHPHQWLQWKH1HZ 3ROLFLHV6FHQDULRUHODWLYHWRWKH&XUUHQW3ROLFLHV6FHQDULR

© OECD/IEA, 2013

The largest eĸciency CO2 emissions savings stem from end-use sectors, parƟcularly in the form of electricity savings in buildings and industry. While electric motor systems account for the greater part of the electricity savings in industry, a phase-out of incandescent light bulbs and stricter appliance standards contribute most to lower electricity consumpƟon in buildings. Eĸciency savings in transport are mainly driven by increased fuel-economy standards for new vehicles, which lead to CO2 emissions savings equal to 12% of total savings in 2035. Eĸciency gains in power plants, transmission and distribuƟon, reĮneries, and oil and gas extracƟon are responsible for about 7% of savings in 2035. A large share of these savings is achieved by reducing the use of ineĸcient coal-Įred power plants and switching to more eĸcient gas-Įred electricity generaƟon (IEA, 2013c).



World Energy Outlook 2013 | Global Energy Trends

Chapter 8 Energy and competitiveness How will price disparities alter global economic geography? Highlights

x DispariƟes in energy prices between countries and regions, especially for natural gas and electricity, have widened signiĮcantly in recent years with implicaƟons for economic compeƟƟveness. Natural gas prices have fallen sharply in the United States, largely as a result of the recent shale gas boom, and today are about one-third of import prices to Europe and one-ĮŌh of those to Japan. Electricity price diīerenƟals are also large, with Japanese and European industrial consumers paying on average more than twice as much for electricity as their counterparts in the United States, and even Chinese industry paying almost double the US level.

x Energy-intensive sectors (chemicals; primary aluminium; cement; iron and steel; pulp and paper; glass and glass products; reĮning) account globally for about 20% of industrial value added, 25% of industrial employment and 70% of industrial energy use. Energy costs can be vital for the internaƟonal compeƟƟveness of energy-intensive industries, parƟcularly where energy accounts for a signiĮcant share of total producƟon costs and where the resulƟng goods are traded extensively. The importance of energy in total producƟon cost is greatest in the chemicals industry, where in some segments it can account for around 80%, including in petrochemicals.

x ShiŌs in industrial compeƟƟveness have knock-on eīects for the rest of the economy. Recent rising energy prices across many regions have resulted in signiĮcant shiŌs in energy and overall trade balances, as well as in energy expenditures taking a growing share of household income. While natural gas price diīerenƟals narrow in our central scenario, gas and industrial electricity prices in the European Union and Japan remain around twice the level of the United States in 2035. The European Union and Japan see a strong decline in their shares of global exports of energy-intensive goods – a combined loss of 30% of their current share – although the European Union sƟll remains the leading exporter. The United States sees a slight increase in its share of exports, with the increase being stronger in many emerging economies, parƟcularly in Asia, where domesƟc demand growth for energy-intensive goods also supports a swiŌ rise in producƟon.

© OECD/IEA, 2013

x High energy prices do not have to result in onerous energy costs for end-users or the naƟonal economy. Improvements in energy eĸciency have a crucial role in miƟgaƟng high energy costs. Policymakers can also boost energy compeƟƟveness by supporƟng indigenous sources of energy supply as well as by increasing compeƟƟon in wholesale and retail energy markets. A carefully conceived internaƟonal climate change agreement can help to ensure that the energy-intensive industries in countries that act decisively to limit greenhouse-gas emissions do not face unequal compeƟƟon from countries that do not. Chapter 8 | Energy and competitiveness

261

Energy and internaƟonal compeƟƟveness The role of energy in internaƟonal compeƟƟveness has become a live issue in poliƟcal, economic and environmental debate around the world in recent years, with the emergence of pronounced dispariƟes in energy prices among countries and regions at a Ɵme of weaker economic growth (Boxථ8.1). The widening of regional price diīerenƟals has accompanied the interplay of a number of important new trends. These include the rebound in United States oil and gas producƟon thanks to the development of shale and other unconvenƟonal energy resources; the opening up of new hydrocarbon provinces in Africa and elsewhere; and a shiŌ in the energy balance away from fossil fuels and nuclear power towards renewable energy sources, notably in the European Union. Box 8.1 ‫' ٲ‬  HÀQLQJ´FRPSHWLWLYHQHVVµ In this chapter, ŝŶƚĞƌŶĂƟŽŶĂů ĐŽŵƉĞƟƟǀĞŶĞƐƐ refers to the ability of both individual Įrms and enƟre economies to compete internaƟonally. As one of several components of the cost of doing business, the price of energy can have a material impact on the cost of producƟon, or producƟvity. Diīerences in energy prices across countries can, therefore, be an important factor in how eīecƟvely Įrms can compete in export markets and with imported goods and services. The term ĞŶĞƌŐLJ ĐŽŵƉĞƟƟǀĞŶĞƐƐ is taken to mean the cost of providing energy services in one economy relaƟve to others. This chapter’s main focus is on the impact of divergences in energy prices on ŝŶĚƵƐƚƌŝĂů ĐŽŵƉĞƟƟǀĞŶĞƐƐ – the ability of industry (parƟcularly its energy-intensive segments) in a given economy to compete internaƟonally. The raƟonale for this is that energy generally accounts for a far bigger share of producƟon costs in industry than in services.

© OECD/IEA, 2013

ĐŽŶŽŵŝĐ ĐŽŵƉĞƟƟǀĞŶĞƐƐ refers to the producƟvity of an enƟre economy relaƟve to others, thus capturing the compeƟƟveness of both industry and services. The producƟvity of an economy determines the level of prosperity that it can aƩain and the rates of return on investments that can be achieved (WEF, 2013a). Higher producƟvity allows naƟonal economies to grow faster over the longer term and sustain higher wage levels, boosƟng the welfare of their populaƟons. The new global energy map that is taking shape has potenƟally important implicaƟons for the relaƟve cost of energy in diīerent countries and, therefore, for the global economic balances established through compeƟƟon between companies operaƟng in diīerent regions. Those countries facing relaƟvely high primary fuel and end-user prices are concerned that the impact on producƟon costs might deter investment and lead to producƟon and jobs migraƟng to countries where energy prices are now signiĮcantly lower, such as the United States. With many countries facing acute economic diĸculƟes, concerns about a loss of compeƟƟveness are moving to centre stage, especially in energy-imporƟng countries, and these could erode internaƟonal and naƟonal eīorts to tackle trade barriers and climate change. Conversely, those countries that are enjoying relaƟvely low energy 262

World Energy Outlook 2013 | Global Energy Trends

prices are hopeful of being able to exploit this advantage by boosƟng producƟon and exports of goods that require signiĮcant amounts of energy. To the extent that lower prices on the internal market result from increased domesƟc energy producƟon, the economy beneĮts from an addiƟonal economic sƟmulus. That is why several countries with large unexploited resources of unconvenƟonal gas are keen to replicate the US experience. The debate about energy and compeƟƟveness has proceeded without much hard data. This chapter seeks to shed light on the quesƟon of how signiĮcant energy is to compeƟƟveness in reality, what persistent energy price dispariƟes might mean for future global economic balances and what policymakers can do to improve economic compeƟƟveness, while at the same Ɵme addressing energy security and environmental concerns. The cost of energy is just one of several factors that aīect the overall cost of producing goods and services, and, therefore, proĮtability. Other costs, including labour, capital, other raw materials and maintenance, also aīect compeƟƟveness signiĮcantly.1 These costs – and the overall aƩracƟveness of an economy to potenƟal investors – are inŇuenced heavily by insƟtuƟonal factors (including Įnancial, monetary, tax, legal and regulatory systems), poliƟcs and geopoliƟcs, infrastructure, technology, educaƟon and labour markets (seeථthe example of
© OECD/IEA, 2013

Despite recent high energy prices, fuel supply and power generaƟon make up just 5% of the global economy today. In general, energy prices play a relaƟvely minor part in the calculaƟon of compeƟƟveness, as in most sectors and countries energy accounts for a relaƟvely small proporƟon of total producƟon costs. But for some types of economic acƟviƟes the share can be much higher, reŇecƟng their degree of energy intensity – the amount of energy needed for each unit of value added. For those acƟviƟes, marked dispariƟes in energy prices across regions can lead to signiĮcant diīerences in operaƟng margins and potenƟal returns on investment, especially where the output is transported over long distances at relaƟvely low cost. In some cases, energy prices can be the single most important factor in determining investment and producƟon decisions. By contrast, the internaƟonal compeƟƟveness of many service acƟviƟes is less aīected by price dispariƟes, as their energy intensity is oŌen low and their output is sold mainly domesƟcally (notable excepƟons include transport services and data centres).

1 2 3 4 5 6 7 8 9 10 11 12 13 14

Persistently high energy price dispariƟes can, therefore, lead to important diīerences in economic structure over Ɵme. Industry (including energy supply) makes up around 30% of world gross domesƟc product (GDP in purchasing power parity ΀PPP΁ terms2), but in

15

1.ഩ IMD World Competitiveness zearbook 2013 ranks the United States highest for overall competitiveness, followed by Switzerland and Hong
17

Chapter 8 | Energy and competitiveness

263

16

18

some countries is much higher or lower. Some regions that are well-endowed with energy resources have always held an energy cost advantage and have, in some cases, developed large export-oriented heavy industries based on low energy prices. Conversely, regions reliant on expensive imported energy – including the United States and much of the European Union – have seen a progressive decline in the share of manufacturing in their economies in recent decades, though other factors have also contributed. This suggests that recent changes in relaƟve energy prices could have far-reaching eīects on investment, producƟon and trade paƩerns. Table 8.1 ‫ ٲ‬6 KDUHRILQGXVWU\LQÀQDOHQHUJ\XVH E\IXHODQGUHJLRQ

OECD Americas

Coal

Oil

Gas

Electricity andථheat

88

20

41

33

Renewables 38

All fuels 32

95

16

42

28

43

27

96

15

39

25

44

26

Europe

75

19

39

37

29

32

Asia Oceania

98

35

37

38

65

45

99

33

25

30

83

41

88

26

59

51

14

45

United States

Japan Non-OECD E.ථEuropeͬEurasia Russia Asia

87

21

42

40

12

41

89

22

46

40

14

42

89

28

65

62

10

51

China

88

27

47

68

ф1

58

India

92

26

91

45

17

42

ASEAN

95

32

92

41

21

41 45

Middle East

90

32

69

22

ф1

Africa

75

13

74

42

10

20

LaƟn America

99

22

73

42

49

41

99

21

82

46

60

44

World

Brazil

88

20

49

43

18

38

European Union

79

20

39

36

30

32

© OECD/IEA, 2013

* If not explicitly menƟoned otherwise in the chapter, energy use within the chemical industry includes petrochemical feedstocks, while the iron and steel sector includes own use and transformaƟon in blast furnaces and coke ovens. Industry energy use does not include power generaƟon or other energy sectors, such as reĮning and hydrocarbon extracƟon, unless otherwise menƟoned.

The importance of industry in a given country can be an indirect indicator of its energy compeƟƟveness, given that energy can account for a signiĮcant share of total input costs to manufacturing. In pracƟce, the share of industry in the overall economy in any given region reŇects a number of factors, of which low energy prices – oŌen based on indigenous resources – is just one. The stage of economic development is important too, as is the role of industry, parƟcularly as energy-intensive heavy manufacturing tends to decline in mature economies, which generally rely more on higher-value manufacturing and services. 264

World Energy Outlook 2013 | Global Energy Trends

Today, the share of industry in total Įnal energy use is highest in developing Asia and among the lowest in the United States (Tableථ8.1). Coal use outside of power generaƟon is parƟcularly concentrated in industry. In the OECD, industry’s share of total Įnal energy use has fallen in recent years as energy use grew in the services and residenƟal sectors. Conversely, it has increased in developing Asia, parƟcularly in China and India. Generally, industrial energy use since 2000 has shiŌed progressively away from oil products towards coal, and electricity and heat (Figureථ8.1). These trends are expected to change over the period to 2035, with electricity and heat and gas gaining market share at the expense of coal and oil in the New Policies Scenario, which assumes the cauƟous implementaƟon of announced policy measures (see Chapter 1).  RUOGHQHUJ\XVHLQLQGXVWU\E\IXHOLQWKH1HZ3ROLFLHV6FHQDULR Figure 8.1 ‫ ٲ‬: 32% 28%

Electricity and heat

24%

Coal Gas

20%

Oil

3 4 5 6

8 9

12% 8%

Renewables

4% 1995

2000

2005

2011

2015

2020

2025

2030

10 11

2035

At any given moment, there can be large diīerences in energy prices across countries, and even within countries according to the precise point of delivery of the fuel and the type of consumer. There are many reasons for these diīerences, the principal ones being diīerences in the cost of transporƟng energy to market, contractual terms governing the way prices are set, taxes, subsidies, labour, other producƟon costs along the energy supply chain, the degree of compeƟƟon in energy markets and trade restricƟons. zet relaƟvely high energy prices do not necessarily result in high energy costs, as they can be miƟgated by eĸcient use of energy and conservaƟon (see Chapter 7). Indeed, high prices strengthen incenƟves to invest in more eĸcient technologies and deter wasteful use of energy. Subsidies that lower the price of energy to users have the opposite eīect (see Chapterථ2). Government policies can therefore play an important role in miƟgaƟng the impact of declining energy compeƟƟveness caused by an increase in relaƟve energy prices. © OECD/IEA, 2013

2

7

16%

1990

1

12 13 14 15 16 17 18

Chapter 8 | Energy and competitiveness

265

S P O T L I G H T What role does energy play in Korea’s industrial success?
© OECD/IEA, 2013


266

World Energy Outlook 2013 | Global Energy Trends

Energy price dispariƟes

1

:ƵƐƚŚŽǁďŝŐĂƌĞƌĞŐŝŽŶĂůĚŝƐƉĂƌŝƟĞƐŝŶĞŶĞƌŐLJƉƌŝĐĞƐ͍ Big diīerences in energy prices paid by consumers in diīerent countries, whether businesses or households, have always existed. But the last few years have seen a substanƟal widening of some of the dispariƟes, notably in natural gas prices between the United States, Europe and Asia. This was mainly as a result of the plunge in wholesale gas prices in the United States due to soaring shale gas producƟon; an increase in oilindexed gas prices in other regions; and higher spot prices for liqueĮed natural gas (LNG) in Asia, largely as a result of the surge in Japanese gas demand that followed the accident at the Fukushima Daiichi nuclear power plant. Smaller diīerences have been observed in prices of reĮned oil products, as they are traded in a highly liquid internaƟonal market and their transport costs are relaƟvely low. However, some oil-producing states (notably in the Middle East) sƟll subsidise oil products heavily, while, on the other hand, many countries have high taxes on oil products. Figure 8.2 ‫ ٲ‬Ratio of Japanese and European natural gas import prices to

175

6

160

5

145

4

130

3

115

2

100

1

85

1Q 03

1Q 05

1Q 07

1Q 09

1Q 11

bcm

7

Rao

190

Japan versus US gas price

4 5 6 7

9

Europe versus US gas price US total gas producon (right axis)

10 11 12 13

70 3Q 13

Notes: The European price is the weighted average price of imports at the German border. The Japanese price is for deliveries of LNG to import terminals. US prices are Henry Hub.

© OECD/IEA, 2013

3

8

United States natural gas spot price 8

2

14

Sources: US EIA, German BAFA (Bundesamt fƺr WirtschaŌ und Ausfuhrkontrolle), and Japanese Ministry of Finance databases; and IEA analysis.

15

The ballooning of wholesale (pre-tax) gas price diīerenƟals between the United States and other regions began in early 2008, but has deŇated since 2012 (Figureථ8.2). By mid-2012, prices in Europe were close to Įve Ɵmes higher than in the United States and prices in Japan were over seven Ɵmes higher. This trend is explained primarily by the surge in US shale gas coupled with a historically mild winter, which has boosted overall gas availability and driven prices down to historically low levels. At the same Ɵme, gas import prices in Europe and the Asia-PaciĮc region, which are mostly indexed to oil prices, have remained high due

16

Chapter 8 | Energy and competitiveness

267

17 18

to sustained high oil prices. A soŌening of internaƟonal oil prices in early 2013 helped drive gas prices outside the United States down somewhat, while gas prices rebounded in the United States as drilling for shale gas fell in areas with high liquids content. The spot price of gas at Henry Hub in the United States doubled from a low of less than $2ථper million BriƟsh thermal unit (MBtu) in April 2012 to $4.2ͬMBtu by April 2013, though it fell back to $3.6ͬMBtu in September 2013. Coal prices can also diīer, both within and across countries, reŇecƟng diīerences in resource endowments, coal quality and the cost of transporƟng coal over land and sea. Recently there have also been major shiŌs in price diīerenƟals between the AtlanƟc and PaciĮc coal markets. During much of the 2000s, steam coal prices in China were at a price discount of around 25-50% to that in Europe (Figureථ8.3). Since 2009, as a result of China becoming and remaining a signiĮcant net importer of coal, the price rose above that in Europe and has since commanded a premium of around 20-50% (see Chapterථ4). These trends have aīected the compeƟƟveness of Chinese industries in coastal regions that rely heavily on coal, relaƟve to their coal-consuming compeƟtors in North America and Europe. Higher coal prices have also created upward pressure on electricity prices in China (though the laƩer remain regulated), as most Chinese power generaƟon is coal-based. In addiƟon, as demand for coal from US power generators had fallen sharply in 2012 due to the low price of compeƟng gas, this led to a surge in US coal exports to Europe (where coal held up due to increased use from generators in response to high gas prices). Figure 8.3 ‫ ٲ‬Ratio of OECD coking to steam coal prices and Asian to

Rao

European steam coal prices 2.1 1.8

Qinhuangdao versus Northwest Europe steam coal prices

1.5

OECD coking versus steam coal prices

1.2 0.9 0.6 0.3

1Q 01

1Q 03

1Q 05

1Q 07

1Q 09

1Q 11

3Q 13

Notes: Qinhuangdao is a major coal port in northeast China. NWE ARA is the northwest Europe marker price for the Amsterdam-RoƩerdam-Antwerp region.

© OECD/IEA, 2013

Sources: McCloskey Coal Report databases and IEA analysis.

RelaƟve to steam coal, the price of coking coal (used primarily in the producƟon of iron) has tended to rise since 2005, due to surging demand and a wave of industry consolidaƟon resulƟng in a high concentraƟon of supply. The premium has diminished since 2011, as 268

World Energy Outlook 2013 | Global Energy Trends

demand from steel producers has slowed and the price of potenƟal high quality subsƟtute steam coal has decreased. This decline in the coking coal price premium has beneĮted steel producers in countries, such as Japan, that rely heavily on coking-coal imports. In response to pressures from large consumers of coking coal, suppliers have recently moved from mostly annual contracts to quarterly contracts, resulƟng in more price volaƟlity. The Įnal prices, including taxes, paid by industry for diīerent fuels vary enormously across countries, especially for electricity (Figureථ8.4). The Middle East has by far the lowest prices for most fuels thanks to low producƟon costs, and, in some cases, large subsidies (seeථChapterථ2). In other regions, the prices for light and heavy fuel oil do not diīer greatly, but the price variaƟons are more pronounced for gas. Regional diīerences in electricity prices reŇect, to some degree, diīerences in the prices of fuels used for power generaƟon: the recent decline in gas prices in the United States has helped reduce electricity prices to a level below that in any other major country outside the Middle East. China’s industrial electricity prices have increased signiĮcantly in recent years, largely because of rising coal prices and cross-subsidies in favour of residenƟal customers. AddiƟonally, even within countries and industrial sub-sectors, the prices paid for energy by industry can vary signiĮcantly.

1 2 3 4 5 6 7 8

Dollars per toe (2012)

Figure 8.4 ‫ ٲ‬,QGXVWULDOHQHUJ\SULFHVLQFOXGLQJWD[E\IXHODQGUHJLRQ 2 500

Japan European Union Brazil China India United States Middle East

2 000 1 500 1 000

9 10 11 12

500

13 Electricity

High sulphur fuel oil

Natural gas

Steam coal

14

Note: toe с tonne of oil equivalent.

© OECD/IEA, 2013

Sources: IEA databases and analysis.

The weighted average price of all fuels in industry, including tax, has increased in most regions in real terms over the past decade, but at diīerent rates (Figureථ8.5). India’s industrial price has risen modestly, partly thanks to subsidies. By contrast in Brazil, the European Union and China, the industrial price more than doubled from 2002 to 2012. Over the same period, Japan’s average industrial price increased less in percentage terms, by 90%, (partly because relaƟvely high taxes dampened the impact of higher internaƟonal energy prices) but it remains one of the highest among the leading economies. The US average industrial price rose by nearly 80% between 2002 and 2008, but it then fell 10% by 2012, and is now one of the lowest amongst the leading economies. Chapter 8 | Energy and competitiveness

269

15 16 17 18

Dollars per toe (2012)

Figure 8.5 ‫ ٲ‬$  YHUDJHLQGXVWULDOHQHUJ\SULFHVLQFOXGLQJWD[E\UHJLRQ 1 200

800

Brazil Japan European Union

600

China

1 000

United States 400

India

200 Middle East 2002

2003

2004

2005

2006

2007

2008

2009

2010

2011

2012

Note: CalculaƟon based on the average price of each fuel (electricity, high sulphur fuel oil, natural gas and steam coal), weighted by the industrial consumpƟon of these fuels. Sources: IEA databases and analysis.

Box 8.2 ‫ ( ٲ‬IIHFWRIWD[HVDQGVXEVLGLHVRQFRPSHWLWLYHQHVV Taxes on the sale of energy to industry aīect the sector’s internaƟonal compeƟƟveness, as taxes push up the eīecƟve price that industries have to pay. But this eīect may be oīset, to some degree, by other government intervenƟons designed to improve industrial compeƟƟveness, so the net eīect can be posiƟve or negaƟve. For example, the revenues raised by those or other taxes may be used to pay for a range of other government measures and programmes, such as improvements to infrastructure and support for investment, that ulƟmately help industry to lower energy and other costs. Judicious use of revenues from relaƟvely high taxes on energy can improve overall economic compeƟƟveness, if it enhances the overall aƩracƟveness of invesƟng in the economy. For example, Switzerland has relaƟvely high rates of energy taxaƟon, yet has come out on top in each of the last Įve years in the World Economic Forum’s annual survey of economic compeƟƟveness among 148 countries (WEF,ථ2013a).

© OECD/IEA, 2013

Energy consumpƟon subsidies may make energy-intensive industries more compeƟƟve, but actually make the overall economy less compeƟƟve. This is because they create market distorƟons that are likely to lead to a misallocaƟon of resources and a resulƟng loss of economic eĸciency and social welfare. Subsidies also weaken prospects for energy eĸciency, as they distort payback period calculaƟons for investments. By encouraging over-consumpƟon, energy subsidies can also give rise to large environmental costs, including emissions higher than would otherwise be the case. One reason why industrial energy prices vary is because of diīerences in rates of taxaƟon. Taxes can aīect industrial compeƟƟveness, but not necessarily economic compeƟƟveness (Boxථ8.2). Sales of energy products to industry generally carry a lower rate of tax than products in the household sector, but they are nonetheless high in some countries –



World Energy Outlook 2013 | Global Energy Trends

notably in Europe. For example, tax accounts for one-third of the total industrial price of electricity in Germany, and 15% in France, while in the United States taxes levied by the states are lower, although the naƟonal average is unknown (Tableථ8.2). Generally taxes are highest on electricity. As in most cases value-added tax is refundable for industry, taxes reported for industry reŇect mainly excise duƟes or other taxes. In some high-tax countries, especially in Europe, tax exempƟons or reducƟons have been proposed, as a way of improving compeƟƟveness.3

1

Energy consumpƟon subsidies also contribute to regional energy price diīerences and they remain signiĮcant in some non-OECD countries, notably among oil exporters (seeථChapterථ2). For example, oil and gas prices to industry in most of the Middle East are far below internaƟonal prices, giving industry in the region a big advantage, but at the same Ɵme they carry large net economic, social and environmental costs. There can also be large cross-subsidies between industrial and household consumers (for example in China), which generally result in higher prices to industry than would be the case if prices were determined according to supply costs.

4

Table 8.2 ‫ ٲ‬6 KDUHRIWD[LQLQGXVWULDOHQHUJ\SULFHVLQVHOHFWHGFRXQWULHV Electricity

Heavy fuel oil

Natural gas

Steam coal

Germany

33

4

10

9

Brazil

26

n.a.

22

n.a.

China

15

20

15

18

France

15

3

4

6

Japan

7

8

6

11

India

n.a.

22

n.a.

16

United States

n.a.

5

n.a.

n.a.

5 6 7 8

10 11

Sources: Sistema Firjan (2013); IEA databases and analysis.

12 13 14

,ŽǁĂƌĞƌĞŐŝŽŶĂůĞŶĞƌŐLJƉƌŝĐĞĚŝƐƉĂƌŝƟĞƐƐĞƚƚŽĞǀŽůǀĞ͍ The tŽƌůĚŶĞƌŐLJKƵƚůŽŽŬ derives energy price trajectories through an iteraƟve modelling process, based on assumpƟons about producƟon costs along the supply curve for each fuel and technology, as well as exisƟng and future contractual terms and other factors that inŇuence both energy supply and demand (see Chapterථ1). The resulƟng regional fossil fuel prices determine the trajectory of end-user prices, taking into account taxes and subsidies (and their eventual phase-out), while electricity and heat prices are derived endogenously. © OECD/IEA, 2013

3

9

Notes: In most cases value-added tax is refundable for EU industry; hence, taxes reported mainly reŇect excise duƟes or other taxes. In Germany, most energy-intensive industries are exempt from the renewables levy and electricity tax, while coal and gas use is also exempt from taxes for most industries. In France and Germany, the tax shares on heavy fuel oil apply to low sulphur fuel oil. Data for China varies depending on product and sector speciĮcaƟon. In the United States, taxes on gas and electricity mostly refer to general sales taxes levied by the states (between 2-6%), although their naƟonal average is unknown; similarly for coal the naƟonal average of various taxes is unknown.

3.ഩ In Germany, energy-intensive industrial sectors benefit from exemptions from the renewables levy, and also some of them are exempt from additional taxes and surcharges. In total, around 15% of electricity sales to German industrial consumers are totally exempt from the renewable levy (BDEW, 2013).

Chapter 8 | Energy and competitiveness

2

271

15 16 17 18

The derived price levels in the New Policies Scenario imply some persistent, large dispariƟes in import and retail prices between regions. These reŇect assumpƟons about taxes and subsidies, as well as the market condiƟons expected to prevail in each region, transport costs between them and supply constraints. There is expected to be liƩle change in the raƟo of reĮned oil product prices across major regions, other than some convergence between regions that currently subsidise oil products and those that do not (on the assumpƟon that subsidies are phased out within the next decade in net oil-imporƟng countries and reduced in net oil-exporƟng countries that have announced plans to do so). Similarly, regional coal prices are expected to move broadly in parallel. By contrast, regional natural gas price diīerenƟals narrow in the New Policies Scenario, but nonetheless remain large in 2035, in part because of the high cost of transporƟng gas over long distances between exporƟng and imporƟng regions. US gas prices in the New Policies Scenario are about half of those in Europe and Japan in 2035 (Figureථ8.6).4 Figure 8.6 ‫ ٲ‬Ratio of Japanese and European natural gas import prices to

Rao

8QLWHG6WDWHVQDWXUDOJDVVSRWSULFHVLQWKH1HZ3ROLFLHV6FHQDULR 7

Japan versus United States

6

Europe versus United States

5 4 3 2 1 1990 1995 2000 2005 2010 2011 2012 2015 2020 2025 2030 2035

© OECD/IEA, 2013

In absolute terms, industrial electricity prices5 are projected to increase in most regions over the KƵƚůŽŽŬ period. This is mainly the result of the evoluƟon of wholesale prices, which increase in line with increasing fossil fuel prices, investment requirements and the eventual pricing of carbon dioxide (CO2) in some countries.6 Japan is an important excepƟon to this trend, as industrial electricity prices are currently very high following the accident at Fukushima Daiichi. They stand at about three Ɵmes those in the United States, 65% higher than in China and 35% higher than in the European Union (Figureථ8.7). Over Ɵme, electricity prices in Japan are expected to move closer to the average of the last decade, falling in the period to the early 2020s and stabilising thereaŌer. This occurs as nuclear power plants gradually resume generaƟng electricity, as assumed in the New 4.ഩ Gas price differentials are about 20% lower on average in a Gas Price Convergence Case (see Chapter 3). 5.ഩ Prices include costs for wholesale (including CO 2); network, retail and other; renewable subsidies; taxes. 6.ഩ Where end-user subsidies persist, they can play a significant role in keeping prices low, while their phase-out can lead to substantial price increases. 272

World Energy Outlook 2013 | Global Energy Trends

Policies Scenario, and renewable energy technologies are deployed, reducing the need for expensive oil- and gas-Įred power generaƟon. Lower demand for imported natural gas unƟl around 2020 also puts downward pressure on internaƟonal prices for these, contribuƟng further to a reducƟon of wholesale electricity prices in Japan. Figure 8.7 ‫ ٲ‬,QGXVWULDOHOHFWULFLW\SULFHVE\UHJLRQDQGFRVWFRPSRQHQWLQWKH

Dollars per MWh (2012)

1HZ3ROLFLHV6FHQDULR 210

Tax

180

Renewables subsidy

150

Network, retail and other

120

Wholesale excluding CO2 price

60 30

Notes: Network, retail and other costs are regulated and, in the case of China, reŇect a cross-subsidy to households that raises the cost to industrial customers. While the United States and China have renewable subsidy schemes, they are partly or fully borne by tax payers rather than reŇected in the electricity tariīs. Sources: IEA analysis and data (including historical data from China’s State Grid Energy Research InsƟtute).

As a basis for comparison, industrial electricity prices in the United States increase throughout the projecƟon period, but remain well below those in the European Union, Japan and China. The increase is due, in roughly equal parts, to rising fossil fuel prices, recovery of investment costs for new power plants, and network expansions and reinforcements (see Chapterථ5). An expanding role for gas-Įred generaƟon and rising gas prices drives up the fuel component of the electricity price, though this is moderated by gains in power plant eĸciencies. Strong deployment of renewables and the emphasis on more-eĸcient convenƟonal technologies push up the investment cost component, which is parƟally oīset by the shiŌ towards less capital-intensive gas-Įred capacity. On the other hand, the increasing share of renewable-based electricity generaƟon reduces industrial electricity prices by lowering electricity wholesale prices through the merit order eīect, while the costs of subsidies to renewables are partly or fully borne by tax-payers rather than reŇected in the electricity tariīs (see Chapterථ6).

© OECD/IEA, 2013

3

5 6 7 8

2012 2020 2035 2012 2020 2035 2012 2020 2035 2012 2020 2035 European Union Japan China United States

Industrial end-user electricity prices in the European Union are currently more than twice those in the United States (Figureථ8.8). The absolute diīerence between EU and US prices increases slightly in the New Policies Scenario mainly due to rising wholesale electricity prices in the European Union. The biggest driver of rising EU wholesale prices is investment cost recovery for new power plants. Currently, cost recovery is very low because of overcapacity in many European countries following the recent economic crisis, combined with a strong

Chapter 8 | Energy and competitiveness

2

4

CO2 price

90

1

273

9 10 11 12 13 14 15 16 17 18

impetus (from support mechanisms) to build addiƟonal renewables capacity. In the New Policies Scenario, the low rate of investment cost recovery is resolved by around 2020, resulƟng in an increase in wholesale prices. This is parƟally oīset by a marked reducƟon in fuel costs, as generaƟon from renewables expands rapidly and displaces fossil-fuelled generaƟon, and more eĸcient power plants are deployed. The CO2 price also contributes (though less so) to increasing wholesale prices. Subsidies to renewables increase unƟl around 2030, adding to rising electricity prices for industry, before falling below today’s levels as the policies that support more expensive technologies expire.  DWLRRI(XURSHDQ8QLRQ-DSDQHVHDQG&KLQHVHWR86LQGXVWULDO Figure 8.8 ‫ ٲ‬5

Rao

HOHFWULFLW\SULFHVLQFOXGLQJWD[LQWKH1HZ3ROLFLHV6FHQDULR 4.0

Versus United States:

3.5

European Union Japan

3.0

China

2.5 2.0 1.5 1.0 0.5 1990

2000

2011

2020

2030

2035

© OECD/IEA, 2013

In China, industrial electricity prices in 2012 were nearly double those in the United States. In absolute terms, this regional disparity grows slightly over Ɵme in the New Policies Scenario. While the wholesale price excluding the costs associated with a CO2 price remains similar to that in the United States through to 2035, the anƟcipated introducƟon of CO2 pricing in China underpins a widening diīerence in the end-user price. The costs associated with a CO2 price are higher in China than elsewhere mainly because of coal’s relaƟvely large share of electricity generaƟon. Conversely, the introducƟon of a CO2 price incenƟvises investment in more eĸcient technologies, limiƟng the increase of the fuel cost component over the projecƟon period and parƟally compensaƟng for the costs associated with a CO2 price. AddiƟonally, China’s industrial electricity price is kept at high levels by the assumed maintenance of a cross-subsidy from industrial to household customers. A reducƟon in network, retail and other costs also helps to miƟgate the impact of growing costs associated with a CO2 price. Similar to the United States, the costs of subsidies to renewables in China are partly or fully borne by tax-payers rather than reŇected in electricity tariīs. The increasing share of renewables in many countries aīects energy compeƟƟveness in several ways. The higher investment costs (as renewables are more capital-intensive than convenƟonal power plants), increased network costs, and cost of subsidies to renewables

274

World Energy Outlook 2013 | Global Energy Trends

(if passed on to end-users) can apply upward pressure on electricity prices. Renewables can also put downward pressure on end-user prices in the short run by reducing wholesale prices through the merit order eīect. In the long run, they can help to reduce reliance on fossil fuels in power generaƟon, cuƫng both total fuel costs and CO2 costs (especially where limits on emissions exist). The net eīect of higher shares of renewables can vary signiĮcantly across countries. It depends criƟcally on whether renewables deployment occurs as part of the regular cycle of investment in new power plants and whether renewables subsidies are paid by end-users.

Energy and industrial compeƟƟveness The extent to which current price dispariƟes are aīecƟng the compeƟƟveness of industry across countries varies according to economic structure. The vulnerability of each sector to an increase in energy prices, relaƟve to that in other regions, depends largely on its energy intensity and the degree to which the manufactured goods are tradable, which in turn depends on the ease and cost of transportaƟon.

toe per tonne

4

6 7

9

United States China European Union Middle East

0.4

3

8

Figure 8.9 ‫ ٲ‬,QGXVWULDOHQHUJ\LQWHQVLW\E\VXEVHFWRUDQGUHJLRQ

0.5

2

5

tŚLJĚŽĞŶĞƌŐLJƉƌŝĐĞĚŝƐƉĂƌŝƟĞƐĂīĞĐƚŝŶĚƵƐƚƌŝĂůĐŽŵƉĞƟƟǀĞŶĞƐƐ͍

0.6

1

Japan

10 11

0.3 0.2

12

0.1

Iron and steel

Petrochemicals

Pulp and paper

13

Cement

Note: Petrochemicals in this graph refers to ethylene producƟon excluding feedstocks.

14

© OECD/IEA, 2013

Source: IEA analysis.

On average, worldwide in 2011, each dollar of industrial value added involved the use of about 135ථgrammes of oil equivalent, with a value of $0.07 or 7%. If industry in one region has to pay 50% more than the world average for its energy, its overall costs will be 3.5% higher (assuming all other producƟon costs are equal). But for some important industrial sectors, energy is a major input to producƟon. In those cases, high relaƟve prices can be a major handicap, parƟcularly where the goods in quesƟon can be transported over long distance easily and at low cost (Aldy and Pizer,ථ2009). The amount of energy needed per tonne of output is generally highest for iron and steel, petrochemicals, and pulp and paper, with variaƟons across countries reŇecƟng mainly diīerences in the processes deployed, the types of products produced and the variaƟons in energy eĸciency (Figureථ8.9). Chapter 8 | Energy and competitiveness

275

15 16 17 18

Worldwide, energy accounts on average for more than one-tenth of total producƟon costs (including labour and capital) in only a handful of industrial sectors, but these sectors account for a relaƟvely large share of total manufacturing value added. The major energyintensive industries are chemicals; primary aluminium; cement; iron and steel; pulp and paper; glass and glass products; as well as reĮning.7 Although energy intensity in some of these sectors, such as aluminium, can be high on average, they account for a relaƟvely low share of overall energy use worldwide (Tableථ8.3). Together, these energy-intensive sectors account for some 20% of total value added by industry and 25% of industrial employment, but 70% of industrial energy use worldwide. The importance of energy in total producƟon cost is greatest in the chemicals industry, where in some segments it accounts for around 80% (Figureථ8.10). In acƟviƟes such as ethylene and nitrogen ferƟliser producƟon, it is actually the fossil fuel used for feedstock that accounts for the bulk of energy costs. Across regions there are big diīerences in the share of energy in total producƟon costs by sector, due to diīerences in energy prices, the cost of other materials, labour and capital, and process eĸciencies. The share is generally highest in Europe and lowest in the Middle East, where energy prices are oŌen heavily subsidised. We esƟmate that the lower price of gas and electricity in 2012 in the United States relaƟve to Europe equated to total savings of around $130ථbillion for the US manufacturing industry. Figure 8.10 ‫ ٲ‬6 KDUHRIHQHUJ\LQWRWDOSURGXFWLRQFRVWVE\VXEVHFWRU Organic chemicals

United States

Nitrogen ferliser

Germany Japan

Primary aluminium Cement Iron and steel Pulp and paper Inorganic chemicals Glass and glass products 20%

40%

60%

80%

100%

Notes: Red horizontal lines show typical ranges for the world. In chemical industries, energy is used both in the producƟon process and as a feedstock. Pulp and paper excludes prinƟng. There are no data for primary aluminium in Japan as producƟon there is minimal.

© OECD/IEA, 2013

Sources: US Department of Commerce (Census Bureau), Eurostat and Federal StaƟsƟcal Oĸce of Germany online databases; Eurostat databases; UNIDO (2010); OECD (2012b); Ecorys, Ğƚ Ăů͘ (2011); Morgan Stanleyථ(2010); IEA esƟmates and analysis.

7.ഩ Some types of oil and gas exploration and production fall into this category as well, notably oil sands and shale oil and gas. 276

World Energy Outlook 2013 | Global Energy Trends

Table 8.3 ‫ ٲ‬,QGLFDWRUVRIVLJQLÀFDQFHRILQGXVWU\WRWKHHFRQRP\E\

1

VXEVHFWRUDQGUHJLRQ

Region Chemicals

Aluminium

Cement

Iron and steel

Pulp and paper

Glass and glass products

ReĮning

Total industry

US Japan EU China World US Japan EU China World US Japan EU China World US Japan EU China World US Japan EU China World US Japan EU China World US Japan EU China World US Japan EU China World

Energy use as share of industry total (%) 36.3 33.2 32.0 19.6 27.8 2.0 0.2 2.3 3.9 2.7 1.4 2.7 4.1 14.1 7.2 6.5 27.5 13.9 35.9 20.0 8.5 4.5 6.9 2.5 3.9 1.4 0.3 1.3 0.9 1.0 15.1 6.2 11.1 3.1 7.5 100 100 100 100 100

Value added Share of Share of GDP industry (%) total (%) 2.3 11.2 2.5 9.3 0.5 2.1 2.4 5.2 2.2 7.2 0.1 0.3 0.1 0.3 0.1 0.2 n.a. n.a. 0.0 0.4 0.04 0.2 0.2 0.9 0.04 0.1 1.2 2.6 0.7 2.3 3.1 0.6 0.9 3.3 0.2 0.6 1.3 2.9 0.7 2.2 0.5 2.5 0.6 2.1 0.3 1.0 0.5 1.2 1.1 3.6 0.2 0.7 0.3 1.0 0.1 0.3 n.a. n.a. 0.1 0.2 0.5 2.6 0.2 0.9 0.1 0.6 0.1 0.3 0.2 0.5 20.7 100 27.3 100 25.4 100 45.5 100 31.2 100

Net trade as % of value added 14 15 155 -7 -4 -49 -146 -611 n.a. 23 -7 3 139 n.a. n.a. -13 80 46 3 -1 12 2 39 10 0 -3 39 25 n.a. -3 -55 -142 -1 247 n.a. n.a. -22 7 -5 4 -1

Employment Share of People industry (thousand) total (%) 700 6.6 340 4.5 1 160 3.9 25 810 12.9 77 930 10.4 50 0.5 10 0.1 90 0.3 n.a. n.a. n.a. n.a. 30 0.2 90 1.1 70 0.2 14 810 7.4 31 940 4.3 290 2.8 220 2.9 560 1.9 15 440 7.7 29 720 4.0 350 3.3 190 2.5 650 2.2 6 790 3.4 29 360 3.9 150 1.4 50 0.7 310 1.0 n.a. n.a. n.a. n.a. 60 0.6 10 0.2 120 0.4 0.8 1 630 4 810 0.6 10 500 100 7 700 100 30 000 100 200 000 100 750 000 100

© OECD/IEA, 2013

Notes: Data for energy use are for 2011, while data for value added, net trade and employment are for 2010 due to data availability constraints. In addiƟon to the sectors listed, the total industry category includes also data for all the less energy-intensive industries. US = United States. EU = European Union. Sources: Databases (including Eurostat; InternaƟonal Labour OrganizaƟon; World Bank and Global Trade Analysis Project; US Department of Commerce, Census Bureau, Annual surveys of manufacturers; NaƟonal Bureau of StaƟsƟcs of China; InternaƟonal Trade Center); US EIA (2013); MICථ(2012, 2013); Eurostatථ(2011); Schmitz, et al. (2012); EAA (2010); ILO (2012); OECD ENV-Linkages model; and IEA analysis.

Chapter 8 | Energy and competitiveness

277

2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18

Figure 8.11 ‫ ٲ‬Share of energy in total material costs in the United States

Cement

Industrial gas

Organic chemicals

Nitrogen ferliser

20%

40%

60%

80%

100%

2005 2008 2011 2005 2008 2011 2005 2008 2011 2005 2008

© OECD/IEA, 2013

Glass and glass product

Pulp and paper

Inorganic chemicals

Primary aluminium

Iron and steel

2011 2005 2008 2011 2005 2008 2011 2005 2008 2011 2005 2008 2011 2005

Electricity

2008

Fuel

2011

Feedstock

Sources: IEA esƟmates and analysis based on US Department of Commerce, Census Bureau databases; Morgan Stanley (2010); OECD (2012).

278

World Energy Outlook 2013 | Global Energy Trends

RelaƟve to ƚŽƚĂůƉƌŽĚƵĐƟŽŶĐŽƐƚƐ, the share of energy in ƚŽƚĂůŵĂƚĞƌŝĂůĐŽƐƚƐ (which make up most of the variable costs for energy-intensive industries) is generally much higher. While the share of energy in total producƟon costs aīects the aƩracƟveness of invesƟng in diīerent regions, the share in total material costs is a more important factor for nearterm producƟon decisions. The wider global spread of best available technologies tends to reduce regional diīerences in eĸciency. Largely due to diverging energy prices, the contribuƟon of energy to total material costs has followed diīering trends in recent years. In most sectors in the United States, including in organic chemicals, the share has fallen since 2005, and especially 2008, due to weaker gas and electricity prices (Figureථ8.11). By contrast, the share of energy to total material costs has remained Ňat or has risen in most cases in Europe and Japan. As diīerences in total producƟon costs oŌen more than oīset the cost of shipping, most energy-intensive industries are characterised by a signiĮcant degree of internaƟonal compeƟƟon, though the extent to which parƟcular products can be traded varies considerably (Tableථ8.4). Generally, chemicals (including organic chemicals and nitrogen ferƟliser), iron and steel, aluminium, and pulp and paper are sectors parƟcularly exposed to internaƟonal compeƟƟon, while cement is the main excepƟon due to its relaƟvely low value as a bulk product, which oŌen renders long-distance transportaƟon uneconomic. Importantly, the migraƟon of steel and chemicals producƟon away from high energy price regions can be limited by the fact that those acƟviƟes are oŌen verƟcally integrated with less energy-intensive and higher value parts of the value chain.

© OECD/IEA, 2013

Although energy represents a small share of total producƟon costs for most industries, regional variaƟons in energy prices can be more marked than variaƟons in costs of other factors. For example, capital costs tend to be similar across regions, as capital competes internaƟonally. The cost of skilled labour can diīer signiĮcantly across regions, but it has tended to converge in recent years as wage rates in the emerging economies have increased and labour markets have become more internaƟonal. Exchange rate movements can also have a major impact on compeƟƟveness. While a stronger currency against the dollar normally lowers the cost of imports of energy and other raw materials, it will raise the price of exported products, which can limit the ability of local Įrms to compete internaƟonally. The decline in the value of the dollar against most other major currencies over the last decade or so has helped to boost the compeƟƟveness of US manufacturing (Figureථ8.12). On the other hand, the relaƟve strength of the euro and yen has exacerbated the problems that Europe and Japan face in trying to improve their compeƟƟveness, although the yen (like the Brazilian real and Indian rupee) has depreciated markedly in recent months. Several signs point to US industry becoming more compeƟƟve relaƟve to the European Union, Japan and some other energy-imporƟng countries (including China), at least in part due to low energy prices. But despite indicaƟons that investments into US manufacturing are starƟng to pickup, it is too early yet for recent global shiŌs in energy compeƟƟveness to show up clearly in manufacturing output and employment staƟsƟcs, because of the planning and investment lead Ɵmes involved. Globally, both manufacturing output and related employment have declined recently in most regions, yet this is mainly because of the global economic downturn and depressed local demand. Views sƟll diīer as to whether a renaissance in US manufacturing is imminent as a result of low energy prices (Boxථ8.3).

Chapter 8 | Energy and competitiveness

279

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18

Table 8.4 ‫ٲ‬

Product characterisƟcs

Use of energy

Degree of internaƟonal compeƟƟon

Four main product categories: base chemicals (e.g. petrochemicals), speciality chemicals, pharmaceuƟcals and consumer chemicals.

Energy intensity varies enormously across products: generally extremely high for base chemicals (up to 80-85% including feedstock) but very low for pharmaceuƟcals.

CompeƟƟve global markets in most of the main chemicals, but trade of some products limited by technical diĸculƟes and economies associated with the integraƟon of processes.

Aluminium Commodity product, but many diīerent Įnal uses aŌer processing.

Electricity only; needs are about 20 Ɵmes higher for primary producƟon than recycling.

Strong compeƟƟon and large traded market due to ease of transport and big regional diīerences in producƟon costs; prices set internaƟonally.

Cement

Standard commodity, with only a few classes of cement; products from diīerent producers can generally be interchanged.

Requires large amounts of primary energy for process heat to produce clinker (from limestone and clay), which is processed into cement. Coal or municipal or industrial waste are oŌen used. Electricity generally used for crushing, grinding, blowers and cooling.

Market is highly internaƟonalised, dominated by a few large mulƟnaƟonal Įrms, but compeƟƟon is generally localised due to the relaƟvely low value of the product by volume (which oŌen renders long distance transport uneconomic).

Iron and steel

Diverse range, including cast iron, crude steel, hot-rolled Įnished products, cold-rolled sheets and plates.

Coking coal is the main fuel in blast furnaces, though use of gas in direct reduced iron is growing worldwide. Electricity used in electric arc furnaces to melt scrap steel and gas is mostly used in steel Įnishing.

CompeƟƟve global market, though most products are traded domesƟcally or regionally, due to high transport costs.

Pulp and paper

Diverse range of raw materials (wood types and waste), products and manufacturing routes.

Highly energy intensive, due to need to heat raw materials and water to dry the pulp and for mechanical and electrical processes. Wood is usually a leading fuel, due to access.

Very compeƟƟve, parƟcularly for newsprint and oĸce paper, due to oŌen large diīerences in raw material and energy costs.

Glass and glass products

QualiƟes and types of product vary; producƟon processes are broadly similar across the world.

Large amounts of energy needed to heat kilns; gas is oŌen the favoured fuel.

Raw materials are heavily traded internaƟonally; trade restricƟons can be large.

ReĮning

LPG, naphtha, gasoline, jet fuel, other kerosene, diesel, fuel oil, petroleum coke and other products.

Up to 10% in total material costs (including feedstock); over 50% excluding feedstock. The more complex the reĮnery (i.e. the more it produces highquality fuels), the higher the energy requirements.

Extremely high, as unit transport costs per tonne are very low.

Chemicals

© OECD/IEA, 2013

7\SRORJ\RIPDLQHQHUJ\LQWHQVLYHLQGXVWULHV

Source: IEA analysis.



World Energy Outlook 2013 | Global Energy Trends

Index (1Q 01 = 1)

Figure 8.12 ‫ ٲ‬Value of the US dollar vis-à-vis other major currencies

1

2.0

2

1.8 1.6 1.4

3

Indian rupee

1.2

4

Brazilian real

1.0 Japanese yen Chinese yuan Euro

0.8 0.6 0.4 1Q 01

1Q 03

1Q 05

1Q 07

1Q 09

1Q 11

6

3Q 13

Source: De Nederlandsche Bank exchange rate online database.

7

Box 8.3 ‫ ٲ‬Reindustrialisation of the US economy: myth or reality? There have been a good many announcements by leading manufacturing Įrms (including General Electric, Ford, Dow, BASF, Voestalpine and Caterpillar) of plans to invest large sums in new plants in the United States, but the jury is sƟll out on whether they signal the beginning of a renaissance for US manufacturing and, if they do, whether low energy prices are the main driver. Indeed the United States has enjoyed several years of relaƟvely low natural gas prices, which has lowered costs for manufacturers. However, outside petrochemicals (and the shale gas industry itself), there is so far liƩle evidence of any resurgence in investment or producƟon. RelaƟve to the six million jobs that disappeared between 2000 and 2009, the contribuƟon to employment for the Ɵme being appears only modest. According to oĸcial government data, over half a million manufacturing jobs have been added since January 2010. But of those, only 50ථ000 have come from overseas Įrms moving to the United States (MorganථStanley,ථ2013).

© OECD/IEA, 2013

zet the logic of a manufacturing renaissance remains compelling, when account is taken of factors beyond the direct boost to income and jobs from increased drilling and the low energy prices that the shale revoluƟon has brought. These factors include the narrowing wage gap between the United States and China and the conƟnuing rise in US producƟvity. In principle, these factors, combined, should help to make US manufacturing more compeƟƟve with other economies, encouraging a shiŌ in producƟon back to the United States, a phenomenon known as ͞reshoring͟. Thanks to the strong mulƟplier eīect of manufacturing jobs, small- and medium-size domesƟcally focused industrial suppliers would beneĮt too. Recent analysis by the Boston ConsulƟng Group points to an imminent surge in US exports of manufactured goods (in part thanks to low energy prices), which together

Chapter 8 | Energy and competitiveness

5

281

8 9 10 11 12 13 14 15 16 17 18

with the eīects of reshoring could add 2.5 to 5ථmillion jobs by 2020.8 By around 2015, the United States is expected to have an export cost advantage of 5-25% over Germany, Italy, France, the United
,ŽǁĚŽĞƐĐĂƌďŽŶƉƌŝĐŝŶŐĂīĞĐƚŝŶĚƵƐƚƌŝĂůĐŽŵƉĞƟƟǀĞŶĞƐƐ͍8 Carbon pricing is not necessarily detrimental to industrial compeƟƟveness: it all depends on how it is implemented and whether similar acƟon is taken in compeƟng economies. In principle, the introducƟon of a carbon price (whether in the form of a tax on fuel use according to its related CO2 emissions or a cap-and-trade system) increases the cost of industrial producƟon insofar as the industry in quesƟon uses fossil fuels. This leads to a risk that carbon-intensive industries in countries that introduce a carbon penalty migrate to other countries that do not, with no net saving in emissions, a phenomenon known as ͞carbon leakage͟. But, in pracƟce, the extent of the increase in costs depends on the level of the carbon price, whether industry is required to pay the enƟre price and whether accompanying measures are introduced to compensate for the higher prices. For example, under the EU Emissions Trading System (ETS), certain energy-intensive industries have been granted free allowances.9 In addiƟon, part or all of the revenue from carbon pricing may be recycled back to energy users in the form of investments towards improved energy eĸciency, or through other, broader supporƟve policies for industry; hence, this may actually increase industrial and energy compeƟƟveness.

© OECD/IEA, 2013

For a given level of carbon price (where there are no free allowances), the potenƟal impact on total material costs is greatest for those industries most reliant on coal and carbon-intensive electricity, relaƟve to the value of their output. In the case of the United States, a hypotheƟcal CO2 price of $10ͬtonne would increase costs on average (over and above current levels) by about 6% for cement; 5% for primary aluminium (due to its heavy reliance on electricity); 2% for pulp and paper; and less than 1% for both iron and steel, and

8.ഩwwǁ͘ďĐŐ͘ĐŽŵͬŵĞĚŝĂͬƉƌĞƐƐƌĞůĞĂƐĞĚĞƚĂŝůƐ͘ĂƐƉdž͍ŝĚсƚĐŵ͗ϭϮͲϭϭϲϯϴϵ͘ 9.ഩ The EU ETS is the world’s largest cap-and-trade system covering all 28 European Union member states plus Norway, Iceland and Liechtenstein. The CO 2 price under the system has fallen in recent years, largely because energy demand has fallen due to recession and there was a large influx of international credits. The price plummeted to less than Φ3ͬtonne in April 2013, following an inconclusive vote by the European Parliament on a plan to delay the introduction of 900 million of the 16 billion tonnes worth of allowances on the market for 2013-2020. It has recovered a little since with a new vote on an amended Commission proposal, which limits the extent to which allowances can be delayed. As of September 2013, the proposal awaits approval by the European Council. 282

World Energy Outlook 2013 | Global Energy Trends

chemicals (Figureථ8.13). Total material costs could increase by signiĮcantly less or could even fall, were the carbon price to be accompanied by addiƟonal measures to encourage or mandate investments in more energy-eĸcient equipment or processes. Figure 8.13 ‫ ٲ‬Sensitivity of US industrial total material costs to CO2SULFHV

2 3

35% Cement

30%

4

Primary aluminium

25% 20%

5

15% Pulp and paper

10%

6

Iron and steel Chemicals

5% 0

20

10

30

7

40 50 Dollars per tonne of CO2

8

Note: A CO2 price is not assumed on petrochemical feedstocks. Source: IEA analysis.

Enhancing industrial compeƟƟveness does not require governments to relegate acƟon to tackle climate change, since climate change poses a far greater threat to naƟonal economies than the adjustments associated with shiŌs between countries in relaƟve energy costs. Properly designed climate change policies can go hand-in-hand with policies to enhance industrial and energy compeƟƟveness. zet the threat of carbon leakage is real and governments need to pursue climate change policies which ensure their domesƟc energy-intensive industries are not penalised by the absence of policy acƟon in other markets. An internaƟonal agreement on climate change, which for example puts a price on carbon, can help to ensure that energy-intensive industries in countries acƟng decisively to limit emissions do not face unequal compeƟƟon from countries that do not.

&ŽĐƵƐŽŶĐŚĞŵŝĐĂůƐ

© OECD/IEA, 2013

1

ϭϬ

The chemicals industry is the biggest industrial consumer of energy worldwide and the energy-intensity of this sector’s various acƟviƟes varies signiĮcantly. Basic petrochemicals (such as propylene and ethylene), as well as inorganic and agricultural chemicals and ferƟlisers, and certain specialty chemical segments (such as industrial gases), are sensiƟve to energy prices – with energy even accounƟng for up to 90% of total material costs in the United States (Figureථ8.14). ProducƟon and investment decisions in these segments are therefore very sensiƟve to regional energy price diīerenƟals, contrary to other acƟviƟes, such as pharmaceuƟcals, for which energy costs are of lower importance. The producƟon of some chemicals, essenƟally bulk petrochemicals and ferƟlisers, in contrast to most other 10.ഩ See Chapter 15 for a discussion of prospects for oil demand in the petrochemical industry.

Chapter 8 | Energy and competitiveness

283

9 10 11 12 13 14 15 16 17 18

industries, relies on energy as both fuel and as feedstock. Most bulk chemicals can be transported economically over long distances, so regions with low energy prices can have a relaƟve cost advantage in their exports. This has prompted concerns in Europe and Asia about the compeƟƟveness of their respecƟve petrochemical industries. Figure 8.14 ‫ ٲ‬Share of energy in total material costs for selected chemical SURGXFWVLQWKH8QLWHG6WDWHV Energy costs

Propylene

Other costs

Butadiene (1,3-) Methanol Ethylene Anhydrous ammonia Benzene Urea Chlorine/causc soda Sodium carbonate Adipic acid 20%

40%

60%

80%

100%

Note: Energy costs include feedstock use. Source: ACC (2013).

© OECD/IEA, 2013

The surge in shale gas in the United States since 2005 has led not just to lower prices for natural gas (methane) but has also boosted the availability and lowered the prices for liqueĮed petroleum gas (LPG) and ethane contained in associated natural gas liquids. This has triggered a wave of planned new investment in US steam crackers and other downstream units in the ethylene supply chain, as well as other faciliƟes for producing propylene, methanol, ammonia, chlorine and other chemical products (Boxථ8.4). Low feedstock costs are conƟnuing to underpin expansions in the Middle East – at $0.75ͬMBtu, gas prices in Saudi Arabia are among the lowest in the world. But in Europe, heavy reliance on relaƟvely expensive naphtha is puƫng ethylene producers at a compeƟƟve disadvantage to both Middle East and US producers, and prompƟng them to consider modiĮcaƟons to their operaƟons (CEFIC, 2013). For example, the company Total is considering a major upgrade of its reĮning and petrochemical complex in Antwerp, involving an increase of diesel producƟon and using LPG as feedstock in the petrochemical unit. In China, which has large deposits of coal and which already produces large amounts of methanol using coal as feedstock, high oil and gas prices are leading companies to increase oleĮns producƟon (including ethylene) using this process (see Chapterථ15, Box 15.4).

284

World Energy Outlook 2013 | Global Energy Trends

Box 8.4 ‫ ٲ‬7 KHUHPDUNDEOHUHQDLVVDQFH of US petrochemicals

1

The slump in gas, ethane and LPG prices in the United States due to the boom in shale gas has given US petrochemical producers a major advantage over many compeƟtors in Europe and other parts of the world that rely primarily on naphtha, an oil-based alternaƟve feedstock. This sharp improvement in the proĮtability of bulk petrochemicals producƟon has boosted uƟlisaƟon rates at exisƟng US plants and led to a surge in plans for new producƟon faciliƟes (Figureථ8.15). Between 2010 and the end of March 2013, almost 100ථchemical industry projects valued at around $72ථbillion were announced (ACC,ථ2013). According to the American Chemistry Council, these investments, were they all to proceed, would boost producƟon capacity by 40% in 2020; provide 1.2ථmillion jobs during the construcƟon phase (to 2020); create over half a million permanent jobs; and give rise to total output worth $200ථbillion per year in the longer term. The majority of the planned projects, many of them for export, involve expansions of capacity for ethylene, ethylene derivaƟves (such as polyethylene and polyvinyl chloride), ammonia, methanol, chlorine, and to some extent for propylene. Roughly half of the announced investments to date are by Įrms based outside the United States. Much of the investment is aimed at making use of the rapidly growing volume of ethane coming onto the US market. However, using solely ethane as feedstock in steam crackers produces just ethylene and almost no other by-products, such as propylene, which may lead to local imbalances in derivaƟve product markets.

Million tonnes

 LVWRULFDODQGSODQQHGHWK\OHQHFDSDFLW\DGGLWLRQVE\UHJLRQ Figure 8.15 ‫ ٲ‬+ 12

Africa

Planned

ASEAN 8

India

6

Middle East

5 6 7 8 9 10

13

United States

14

2

2010

2011

2012

2013

2014

2015

2016

15

2017

Sources: ICIS (2013); IHS (2013); METI (2013); PlaƩs (2013); US EIA (2013); and IEA analysis.

16

&ŽĐƵƐŽŶŝƌŽŶĂŶĚƐƚĞĞů © OECD/IEA, 2013

4

12

China 4

3

11

Russia

10

2

Iron and steel producƟon requires large amounts of energy and in 2011 the sector accounted globally for 20% of industrial energy use and 8% of total Įnal energy use. Energy typically makes up 10% to 40% of total producƟon costs and therefore the economics of iron and steel producƟon are highly sensiƟve to local energy prices (see EU example in Chapter 8 | Energy and competitiveness

285

17 18

Boxථ8.5). However, the cost of long-distance transportaƟon of Įnished steel products provides a cushion against compeƟƟon from producers in low energy-price regions, and for some specialised products, high quality can compensate for diīerences in producƟon cost. Box 8.5 ‫[( ٲ‬SHQVLYHHQHUJ\DGGVWRWKHVWHHOZRHVRIWKH(XURSHDQ8QLRQ High energy prices are contribuƟng to the diĸculƟes faced by steel producers in the European Union, where domesƟc demand has fallen due to the region’s economic slowdown. EU producers have a strong compeƟƟve posiƟon in domesƟc markets, parƟcularly in high value-added products, and have established strong technological links with their main client sectors (like the automoƟve, aerospace and highperformance engineering industries) to develop tailor-made products. Indeed, EU steel imports have fallen and exports risen since 2009. But several steel plants, in response to lower demand and cost pressures, have closed temporarily or permanently over the past year. According to the European Steel AssociaƟon, EU steel consumpƟon is expected to drop by 4.4% in 2013, before recovering slightly in 2014 (Eurofer, 2013).

© OECD/IEA, 2013

Worryingly for EU steel producers, there appears to be only limited potenƟal for lowering the energy intensity of producƟon through the adopƟon of best available technologies, though innovaƟve technologies under development could yield much bigger gains (Moya and Pardo, 2013). One such process is HIsarna, in which iron ore is processed almost directly into steel, promising much greater energy eĸciency, as well as lower CO2 emissions. A pilot unit for HIsarna is under construcƟon at the Tata Steel plant in IJmuiden in the Netherlands. In June 2013, the European Commission released a plan for the steel industry, which proposes a number of acƟons to help alleviate the problems facing the EU steel industry (EC, 2013). These include moves to ensure EU steel producers have access to third-country markets through fair-trade pracƟce; streamlining EU regulaƟon; promoƟng innovaƟon, energy eĸciency and sustainable producƟon processes; and targeted measures to support employment in the sector and during the restructuring to ensure that highly skilled labour is retained in Europe. The way steel is made is changing in some parts of the world, in part due to shiŌs in the price diīerenƟals between the fuels that can be used in the producƟon process. The standard blast furnace route to making steel involves the use of coking coal, along with iron ore and limestone to produce iron, which is then fed into a basic oxygen furnace (usually together with scrap steel) to produce crude steel. The other main route is the electric arc furnace, which relies on electricity to melt the steel (usually scrap) before further processing. Due to the rising cost of coking coal, some steel producers are turning to an alternaƟve method for producing iron – direct reduced iron (DRI) – which involves the use of a gas (a mixture of hydrogen and carbon monoxide) as a reducing agent (usually derived from natural gas or coal). This process has the advantage of being less capital intensive than the blast furnace method and less carbon intensive, if based on gas. India and Iran (where gas prices are relaƟvely low despite recent energy subsidy reforms) dominate DRI producƟon today, but several plants are under construcƟon in other countries. For example the US steel 286

World Energy Outlook 2013 | Global Energy Trends

Įrm Nucor is expected to bring online a 2.5ථmillion tonnesͬyear gas-based DRI plant in Louisiana at the end of 2013, while the Austrian steel producer, Voestalpine, announced in March 2013 that it would also build a similar 2ථmillion tonnesͬyear plant in Texas (the produced iron will be shipped to Austria for processing into steel).

1

&ŽĐƵƐŽŶƌĞĮŶŝŶŐϭϭ

3

The global reĮning industry is undergoing a profound transformaƟon as a result of changes in regional demand trends for oil products and in feedstock composiƟon, as well as diverging regional energy costs. Crude oil and part of natural gas liquids are mainly used as feedstock in reĮning and to provide fuel for the transformaƟon process (up to 10% of the energy contained in the feedstock). Consequently, the cost of energy inputs has an impact on proĮtability. In general, it is more economical to ship crude oil to reĮneries located close to market than to transport reĮned products over long distances (as separate carriers are needed to ship ͞clean͟ and ͞dirty͟ products), which provides a degree of protecƟon for reĮneries against distant compeƟtors. Nonetheless, imbalances between local producƟon and demand usually mean that signiĮcant volumes of speciĮc products need to be imported or exported. RelaƟvely high energy costs, alongside falling demand and overcapacity, are contribuƟng to weak margins in some regions, notably Europe, where several reĮneries have already closed in recent years and further closures are likely. The importance of low energy prices has risen in the reĮning industry, as the energy intensity of the sector has increased since the mid-2000s. Several factors are contribuƟng to growing energy intensity, notably increasingly stringent oil-product standards (such as low-sulphur diesel) and a combinaƟon of increasing demand for middle disƟllates (diesel and kerosene) coupled with a growth of the share of both very heavy and light oil producƟon, which, together, are forcing reĮners to increase secondary processing. These factors are outweighing improvements in the energy eĸciency of reĮning operaƟons from new investment in energy-saving equipment and improved operaƟng pracƟces. Worldwide, reĮnery gas and oil products (ordinarily produced by the reĮnery itself) are the principal sources of the energy consumed in the reĮning process, though natural gas is a key fuel in regions where gas prices are low, such as the United States. The fall in gas prices in the United States has given its reĮners a compeƟƟve boost, especially relaƟve to European reĮners that are burdened with high imported gas costs, and reĮners elsewhere that rely heavily on oil products to fuel their plants.

© OECD/IEA, 2013

The outlook for industrial energy and compeƟƟveness Industrial energy use over 1987-2011 expanded by almost 55%, and it is projected to increase a further 37% by 2035 in the New Policies Scenario. Gas, electricity and heat increase their combined share of the industrial fuel mix from 43% in 2011 to 50% in 2035, as these fuels account for nearly 70% of the incremental energy demand. Among the energy-

2

4 5 6 7 8 9 10 11 12 13 14 15 16 17 18

11.ഩ See Chapterථ16 for a detailed analysis of the outlook for global refining.

Chapter 8 | Energy and competitiveness

287

intensive industries, some see a marked slowdown in the growth of energy use largely due to lower output growth, as well as process changes and energy eĸciency improvement (Figureථ8.16). The reducƟon is parƟcularly marked in the case of cement, and iron and steel, as the construcƟon boom slows in China. Conversely, energy use in the chemical, and pulp and paper industries expands in absolute terms similarly to historical trends. The chemical industry alone accounts for 40% of incremental industrial gas consumpƟon. The increase in chemicals producƟon is parƟcularly driven by petrochemicals, where the demand for plasƟcs increases strongly in China and other developing Asian countries as a result of their per capita consumpƟon currently being about one-fourth of the OECD level. Overall the share of the four energy-intensive sectors in total industrial energy demand declines from 65% in 2011 to 58% in 2035. The combined energy use of all other industrial sectors almost doubles, driven by increasing producƟon in sectors such as texƟles, car manufacturing, machinery and mining. Figure 8.16 ‫ ٲ‬:  RUOGLQFUHPHQWDOHQHUJ\GHPDQGE\LQGXVWULDOVXEVHFWRU

Cement

Pulp and paper

Iron and Chemicals steel

DQGIXHOLQWKH1HZ3ROLFLHV6FHQDULR Coal 1987-2011

Gas Electricity and heat

1987-2011 2011-2035

Renewables

1987-2011 2011-2035 1987-2011 2011-2035 -75

© OECD/IEA, 2013

Oil

2011-2035

0

75

150

225

300

375

450

525 Mtoe

The energy use, producƟon and export prospects for the energy-intensive industrial sectors diīer markedly between regions. Their stage of economic development is the main determining factor, though energy prices also weigh in. The largest increases in energy demand in the chemical industry come from China, the Middle East and ASEAN countries, while in the iron and steel industry India sees the largest absolute growth. In many emerging economies, the strong growth in domesƟc demand for energy-intensive goods supports a swiŌ rise in their producƟon (accompanied by export expansion). But relaƟve energy costs play a more decisive role in shaping developments elsewhere, parƟcularly among the OECD countries. While regional diīerences in natural gas prices narrow in our central scenario, they nonetheless remain large through to 2035, and electricity price diīerenƟals largely persist.

288

World Energy Outlook 2013 | Global Energy Trends

OECD countries currently dominate the export market for energy-intensive goods, accounƟng for more than two-thirds of the export value. About half of those exports originate from within the European Union, which makes it the largest export region (Figureථ8.17). In the New Policies Scenario, export growth rates among OECD countries are highest in the United States, which enables it to increase its export market share.12 Despite a conƟnuing slow expansion of export volumes, market shares fall in Japan and in the European Union – especially for chemicals. Next to high energy prices, relaƟvely high wages in the European Union as well as longer shipment distances to the major consumpƟon centres in Asia (which emerge in the long term), put European Union exports at a comparaƟve disadvantage. Despite a reduced share in the global export market, which is parƟcularly pronounced up to 2020, the European Union sƟll remains the leading exporter of energy-intensive goods. In 2035 the European Union is exporƟng more than the United States, China and Japan combined. Figure 8.17 ‫ ٲ‬5  HJLRQDOVKDUHVRIJOREDOH[SRUWPDUNHWYDOXHRIHQHUJ\

Other E. Europe/Eurasia Middle East

India

Rest of OECD

United States

12

European Union 2011

2020

2035

Notes: Energy-intensive industries covers chemicals; iron and steel; pulp and paper; cement; and nonferrous metals (aluminium, copper, lead, nickel, Ɵn, Ɵtanium, zinc and alloys such as brass). Intra European Union trade Ňows are excluded. Sources: OECD ENV-Linkages model and IEA analysis.

© OECD/IEA, 2013

The growth in exports from developing Asian countries, including China and India, remains rapid in the New Policies Scenario on the back of rising producƟon of chemicals, aluminium and steel (in some cases). As a consequence, developing Asia increases its export market share to a level equal to that of the European Union. In China, the increase in export market share occurs largely within this decade as steel producƟon is anƟcipated to level oī aŌerwards. Developments in the chemical and non-ferrous industries (where energy

12.ഩ These projections are sensitive to assumptions about real exchange rates, which remain constant through the projection period.

Chapter 8 | Energy and competitiveness

6

11

Japan 20%

5

10

China 40%

4

9

Other developing Asia 60%

3

8

Rest of non-OECD

80%

2

7

LQWHQVLYHLQGXVWULHVLQWKH1HZ3ROLFLHV6FHQDULR 100%

1

289

13 14 15 16 17 18

accounts for the largest share in total producƟon costs) provide a clear indicaƟon of the link between relaƟvely low energy prices and internaƟonal compeƟƟveness. Chemicals producƟon in most regions grows in the New Policies Scenario, but prospects for exports in major producing countries varies markedly (Figureථ8.18a). By 2035 the European Union retains the largest share in the global export market for chemicals (as speciality chemicals, with lower energy-intensiƟes, account for a growing share of global chemicals trade), but its dominance declines by ten percentage points over the projecƟon period. Japan, where petrochemicals account today for around 60% of its chemical exports, sees a marked decline in export market share as petrochemicals producƟon wanes. By contrast, the market shares of China, India and the Middle East increase, supported by strong growth in export values. In 2035, the United States maintains its posiƟon as the second-largest exporter of chemicals in the world, supported by relaƟvely low gas prices and increasing producƟon of bulk chemicals. The narrowing of gas price diīerenƟals in the later part of the projecƟon period boosts the rate of growth in EU exports and tempers the decline in the region’s global market share.

© OECD/IEA, 2013

In the non-ferrous metals industry, the Middle East sees a strong growth in exports and by 2035 becomes the dominant exporter supported by relaƟvely low electricity prices (Figureථ8.18b). The European Union loses its leading trade role, experiencing a nine percentage points drop in global export market share, about equivalent to the gain in the Middle East. While China maintains its share in export markets (the rising producƟon mainly saƟsĮes domesƟc needs), the United States increases its share slightly and Japan’s share decreases slightly. The non-ferrous metal sector is dominated by aluminium and other base metals, such as copper, zinc, and lead, with the rest being made up by precious metals (e.g. gold and sliver) and specialty metals (e.g.ථcobalt). ParƟcularly for primary aluminium, energy costs far outweigh other costs, such as labour, capital or administraƟve costs. This puts regions with low electricity prices, such as the Middle East, Norway or Iceland, at a compeƟƟve advantage. Access to raw material plays a role in other segments: Chile, for example, is the leading exporter of copper thanks to its vast reserves. In the iron and steel industry, the outlook for internaƟonal trade is broadly similar to other energy-intensive sectors. Today, the industry is dominated by China, which represents almost half of current steel producƟon, followed by the European Union with 12%, Japan with 7%, the United States with 6% and Russia and India with each 5%. The vast majority of Chinese steel is used in its domesƟc construcƟon industry, driven by the need for housing in its rapidly developing ciƟes. Only a small part of domesƟc producƟon is currently shipped from China to other countries: China accounts for less than 10% of the global export market. With the domesƟc construcƟon boom slowing notably down towards the end of this decade in the New Policies Scenario, China is able to increase its share in global export markets though from a low base. In light of the structural shiŌs in steel producƟon in mature markets, the OECD sees a drop in market share, with Europe losing the most.



World Energy Outlook 2013 | Global Energy Trends

Figure 8.18 ‫ ٲ‬5  HJLRQDOVKDUHVRIJOREDOH[SRUWPDUNHWDQGJURZWKLQH[SRUW YDOXHVE\VHOHFWHGVHFWRULQWKH1HZ3ROLFLHV6FHQDULR

2

40%

8%

30%

6%

20%

4%

10%

2%

5

0%

6

European Union

United States

China

India

Middle East

Japan

Export value growth

Global export market share

(a) Chemicals 2011

3

2035 CAAGR 2011-2035 (right axis)

4

-2%

7

(b) Non-ferrous metals

8

20%

8%

15%

6%

10%

4%

5%

2%

Export value growth

Global export market share

1

2011

9

2035 CAAGR 2011-2035 (right axis)

10 11

0%

12

-2% Middle East

China

European Union

United States

India

Japan

13

Notes: CAAGR is compound average annual growth rate. Chemicals include base chemicals (e.g. petrochemicals), specialty chemicals, pharmaceuƟcals and consumer chemicals. Non-ferrous metals include aluminium, copper, lead, nickel, Ɵn, Ɵtanium, zinc and alloys such as brass. Intra European Union trade Ňows are excluded. Sources: OECD ENV-Linkages model and IEA analysis.

© OECD/IEA, 2013

&ŽĐƵƐŽŶƚŚĞŽƵƚůŽŽŬĨŽƌĐŚĞŵŝĐĂůƐϭϯ Chemicals are a key energy-intensive industrial sector in the economies of China, the European Union, Japan and the United States, many of which are net exporters of chemicals. The chemical industry is very diverse in terms of output, but energy consumpƟon is dominated by a few large-volume products. OleĮns producƟon, including ethylene and propylene, and their derivaƟves (e.g. polyethylene and ethylene oxide), 13.ഩ See Chapter 15 for a discussion of prospects for oil demand in the petrochemical industry.

Chapter 8 | Energy and competitiveness

291

14 15 16 17 18

make up the largest share of fuel and feedstock use within the chemical industry. Other important intermediate products are aromaƟcs, nitrogen ferƟliser and methanol. Globally, energy use (including feedstock) in chemicals producƟon grows on average 1.5% per year between 2011 and 2035 in the New Policies Scenario, with nearly 70% of the growth met by gas and oil. The chemical industry alone accounts for 35% of incremental industrial energy consumpƟon. The projected increase in chemical energy use in absolute terms and growth rate varies markedly across regions, mainly according to the rate of increase in domesƟc demand, but also as a reŇecƟon of the internaƟonal compeƟƟveness of domesƟc producƟon (Figureථ8.19). Figure 8.19 ‫ ٲ‬Compound average annual change in chemicals energy use DQGSURGXFWLRQE\UHJLRQLQWKH1HZ3ROLFLHV6FHQDULR China

Producon

2011-20 2020-35

Energy use

2011-20 India 2020-35 Middle 2011-20 East 2020-35 United 2011-20 States 2020-35 European 2011-20 Union 2020-35 Japan

2011-20 2020-35 -2%

0%

2%

4%

6%

8%

China and the Middle East alone account for 70% of incremental energy use in the chemical industry to 2035. Energy use falls in the European Union and Japan, but grows in most other major regions, mainly as a result of diīerent rates of growth in the producƟon of chemicals. In the United States, energy needs grow relaƟvely slowly over the enƟre projecƟon period, but this hides a signiĮcant increase in producƟon out to 2020 (supported by a surge in ethane availability), and a fall towards the end of the projecƟon period. The contrast in chemical industry trends between the United States (where output and related energy use grow), and the European Union and Japan (where output and related energy needs decline) is parƟcularly striking and illustrates the central role energy prices can play in industrial compeƟƟveness (although other factors, such as weak domesƟc demand, are important too).

© OECD/IEA, 2013

Energy and economic compeƟƟveness A change in relaƟve energy costs across countries not only aīects industrial and energy compeƟƟveness but also economic compeƟƟveness. The extent to which an increase,

292

World Energy Outlook 2013 | Global Energy Trends

relaƟve to other economies, in the pre-tax price of energy (rather than simply higher prices) undermines economic compeƟƟveness depends largely on the extent to which a given country relies on energy-intensive manufacturing, as well as the scope for higher prices to be oīset by economically viable investments towards greater energy eĸciency. A loss of economic compeƟƟveness (to a greater or lesser degree due to a rise in relaƟve energy costs) may result in a reallocaƟon of resources away from energy-intensive industries towards less energy-intensive manufacturing or services. Conversely, a fall in relaƟve energy costs boosts economic compeƟƟveness. While energy-intensive industries across regions are directly aīected by any change in relaƟve energy prices, this impact on industry also has knock-on eīects more broadly. CompeƟƟvely priced industrial goods (such as cement and steel) help to lower the cost of producing Įnal products (such as housing and metal goods). Also, increasing domesƟc producƟon of energy, oŌen associated with lower energy prices, enhances economic acƟvity indirectly through increased demand for equipment, materials and services (such as steel products, cement, haulage and engineering). In the United States, this eīect may be greater than the beneĮt from lower industrial energy prices (Boxථ8.3). Conversely, a rise in the price of industrial goods, due to high energy prices, erodes indirectly the compeƟƟveness of other sectors through the same mechanisms. Economic restructuring that results from a change in industrial compeƟƟveness and, therefore, economic compeƟƟveness (whether resulƟng from higher energy costs or an increase in the cost of other inputs to producƟon) is generally associated with mediumterm adjustment eīects (such as a change in employment levels, corporate proĮtability, real wages and the rate of inŇaƟon). Any loss of compeƟƟveness is reŇected in a relaƟve decline in GDP when higher-value manufacturing shiŌs to lower energy-cost regions. Also as real disposable incomes falls due to the increase in the share of energy in total household spending, this reduces the amount of money available for spending on other goods and services. MulƟplier eīects accentuate these macroeconomic worries. Conversely, countries with relaƟvely low energy prices enjoy a macroeconomic boost from increased investment, higher incomes and an improvement in their trade balance.

© OECD/IEA, 2013

The global economic rebalancing that follows a shiŌ in energy compeƟƟveness also involves second-order eīects that may temper the eīects of the iniƟal adjustments. For example, although low US gas prices are leading to a loss of energy compeƟƟveness ǀŝƐͲáͲǀŝƐ the United States in higher cost regions, part of this negaƟve impact is being oīset by increased US imports of other products. The economic beneĮts of a fall in relaƟve energy costs due to greater exploitaƟon of domesƟc resources can be reduced by an accompanying sharp inŇow of foreign currency – a phenomenon known as ͞Dutch disease͟. The currency inŇows lead to currency appreciaƟon, which makes the country’s non-energy goods and services less price compeƟƟve on the export market. This can then lead to higher levels of cheaper imports, oīseƫng the impact of lower energy prices on energy compeƟƟveness.

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18

Chapter 8 | Energy and competitiveness

293

tŚĂƚŝƐƚŚĞŝŵƉĂĐƚŽĨĞŶĞƌŐLJƉƌŝĐĞĚŝƐƉĂƌŝƟĞƐŽŶŽǀĞƌĂůůƚƌĂĚĞďĂůĂŶĐĞƐ͍ The divergence in energy prices across regions and ŇuctuaƟons in internaƟonal energy prices in recent years have manifested themselves in shiŌs in energy trade balances. Annual spending on energy imports in 2012 hit new records in dollar terms (by exceeding previous peaks in 2008) in many major energy-imporƟng regions. The contrast between the United States and other major importers is striking; the United States saw its energy import bill fall by 40% since 2008, while that of the European Union slightly increased and that of many others conƟnued to climb. Such ŇuctuaƟons in energy trade balances have been an important driver of recent changes in overall trade balances of the major energyimporƟng regions (Figureථ8.20). This is most evident in Japan, where a sharp increase in energy import costs was the primary cause of the country recording an overall trade deĮcit in 2011. Japan ran its ĮŌeenth straight monthly trade deĮcit in September 2013, making it the longest period of deĮcit since the fourteen months between July 1979 and August 1980. Energy now accounts for one-third of Japan’s total imports, which is a slightly lower share than at the previous peak in 2008. The overall trade deĮcit in the United States worsened steadily over 2009-2011, but a drop in the share of energy in total imports in 2012 helped to reverse this upward trend.14 In the European Union the overall trade deĮcit turned posiƟve in 2012, as strong growth in non-energy exports (parƟcularly from Germany) outweighed the increase in the weight of energy in total imports.

Billion dollars (2012)

 YHUDOOWUDGHEDODQFHLQFOXGLQJHQHUJ\WUDGHE\UHJLRQ Figure 8.20 ‫ ٲ‬2 600

60%

Energy trade balance

400

40%

Non-energy trade balance

200

20%

Net trade balance Energy as share of total imported goods (right axis)

0 -200 -400

2008 2009 2010 2011 2012

2008 2009 2010 2011 2012

2008 2009 2010 2011 2012

-800

2008 2009 2010 2011 2012

-600

© OECD/IEA, 2013

United States Japan European Union China Sources: MIC (2012); WTO (2012) and WTO databases; and IEA analysis.

14.ഩ The US administration deems unhealthy a world economy that is too dependent on US consumption spending and aims to reduce further the US trade deficit. The US National Export Initiative is intended to increase US exports to help reduce worldwide trade imbalances (White House, 2013). 294

World Energy Outlook 2013 | Global Energy Trends

Generally, oil accounts for the majority of fossil-fuel import bills in energy-imporƟng countries, though natural gas import bills can also be signiĮcant (with a share of around 25% in both Japan and the European Union), especially given regional gas price dispariƟes. China in 2012 had a record high fossil fuel net import bill of $270ථbillion (or 2.1% of GDP) represenƟng a four-fold increase over 2005 (Figureථ8.21). Japan’s fossil fuel net import bill rose to over 6% of GDP in 2012, mainly because of a combinaƟon of higher prices and higher imports of energy to replace the loss of nuclear power. In the New Policies Scenario, spending on fossil fuel net imports conƟnues to rise strongly in China and India, resulƟng in China surpassing the European Union spending levels by 2035. The share of GDP spent on fossil fuel net imports declines progressively in all regions, mainly because of eĸciency gains reducing the need for imports. The share falls most in Japan due to a gradual resumpƟon of nuclear power generaƟon and greater generaƟon from renewables, coupled with a push for energy eĸciency improvements. Figure 8.21 ‫ ) ٲ‬RVVLOIXHOQHWLPSRUWELOOVE\UHJLRQLQWKH1HZ3ROLFLHV

1.4% 0.5%

1.2%

0.7%

2.0% 2.1%

1.2% India 100

2.8% 2.8% 200

6

8

10

2035 % Share of GDP (PPP)

11

4.5% 3.6%

China

5

2020

5.8% 6.2%

Japan

4

9

2012 3.3% 3.2% 2.8% 2.3%

2.4%

3

2011

1.5%

European Union

© OECD/IEA, 2013

2.0%

1990-2008 average

2.4%

2

7

Scenario United States

1

1.9%

12 1.6%

13 2.6%

2.2% 300

400

500

14

600 700 800 Billion dollars (2012)

A persistent trade deĮcit can consƟtute a drag on economic growth, manufacturing acƟvity and employment, as each dollar spent on imports that is not matched by a dollar of exports reduces overall demand within an economy. G-20 economies were parƟcularly aīected by the slowdown in internaƟonal trade during the recent economic crisis (OECDͬILOͬ World BankͬWTO, 2010). In the longer term, deterioraƟon in the terms of trade for energy would be expected to lead to currency depreciaƟon, discouraging imports and encouraging exports of goods and services. This would lead towards a narrowing, if not eliminaƟon, of the overall trade deĮcit.

Chapter 8 | Energy and competitiveness

295

15 16 17 18

tŚĂƚŝƐƚŚĞŝŵƉĂĐƚŽĨĞŶĞƌŐLJƉƌŝĐĞĚŝƐƉĂƌŝƟĞƐŽŶŚŽƵƐĞŚŽůĚŝŶĐŽŵĞ͍ Rising energy prices in recent years, combined with the relaƟvely low short-term price elasƟcity of household energy demand, have resulted in energy taking a growing share of household income in most regions (Figureථ8.22). In general, the increased burden on household income is due to higher energy use and prices. The share of energy in EU household income reached a high of almost 8% in 2008, reŇecƟng the important price increase of transport fuel and higher household prices for natural gas and electricity, parƟally driven by increasing taxes. The share is already slightly lower in 2011 and declines by another third to 2035, driven by signiĮcant eĸciency improvements in personal transport, though it remains the highest among leading economies. Figure 8.22 ‫ ٲ‬6 KDUHRIHQHUJ\H[SHQGLWXUHVLQKRXVHKROGLQFRPHE\UHJLRQ LQWKH1HZ3ROLFLHV6FHQDULR 8%

European Union

7%

China

6%

India

5%

United States

4%

Japan

3% 2% 1% 2000

2005

2008

2011

2020

2035

Note: Excludes upfront energy investment costs (e.g. purchase of vehicle). Sources: United NaƟons and Eurostat databases; and IEA analysis.

© OECD/IEA, 2013

In the United States, the share of energy in household income is lower than in the European Union due to low taxes, lower prices for gas and electricity, and higher income levels. While residenƟal energy expenditures (including for space heaƟng, appliances or cooking) were higher than transport-related expenditures in 2000, rising costs for gasoline had reversed this situaƟon by 2011. Personal transport plays an important role in US households’ energy expenditures given the lower use of public transport and larger vehicles on average compared with most other OECD countries. In the New Policies Scenario, the adopƟon of more eĸcient cars in the United States leads to a signiĮcant reducƟon in energy expenditures relaƟve to income by 2035, with Japan following a similar trajectory. In non-OECD countries, including China and India, the share of energy spending is currently below the average of OECD countries as a consequence of the signiĮcantly lower level of cars per capita and lower ownership of household appliances. In China, the share of energy expenditures in household income increased rapidly over the past eleven years not only

296

World Energy Outlook 2013 | Global Energy Trends

as a result of increasing fossil fuel prices but also improved living standards, accompanied by higher energy demand for electrical appliances and space cooling. The importance of energy in income of Chinese households remains roughly stable to 2035, with eĸciency gains moderaƟng the increase in energy demand. In India, the role of energy in household income increases as a consequence of the assumed subsidy phase-out for natural gas and electricity, increasing access to energy and higher use of personal transport.

ŶĞƌŐLJĐŽŵƉĞƟƟǀĞŶĞƐƐĂŶĚƉŽůŝĐLJŝŵƉůŝĐĂƟŽŶƐ There is considerable scope for acƟon to enhance energy compeƟƟveness, both by minimising energy prices and by miƟgaƟng the impact of price increases. It is for businesses and households themselves to make the investments needed and to adjust their spending to respond to changes in the global energy landscape. But it is up to policymakers to create the condiƟons that encourage businesses and households to take the necessary acƟon, so aiding Įrms to compete internaƟonally and for households to obtain aīordable energy services. For example, there has been considerable recent debate about the vulnerability of the European Union’s industrial sector to relaƟvely high energy prices (Boxථ8.6). The challenge for all governments is to idenƟfy win-win soluƟons that improve energy compeƟƟveness (or at least miƟgate part of the impact of energy price dispariƟes), while at the same Ɵme addressing energy security and environmental concerns. Should it not be possible to Įnd ways to compensate for the relaƟve energy price dispariƟes, it would be advisable for policymakers not to impede the economic restructuring that is necessary to respond to shiŌs in energy compeƟƟveness. Without market distorƟons, it oŌen makes economic sense for highly energy-intensive acƟviƟes to migrate to countries that have low energy prices, and for relaƟvely high energy-price countries to focus more on less energyintensive and higher-value-added acƟviƟes. One way to cut energy prices to end-users is to lower taxes, but this is unlikely to make any diīerence to the overall burden of energy on the economy and would counteract eīorts to curb energy imports and reduce emissions. Similarly, introducing subsidies might enhance industrial compeƟƟveness in the near term but in the long term they create large economic, social and environmental costs. Hence other more economically and environmentally eĸcient ways to enhance energy compeƟƟveness should be sought.

© OECD/IEA, 2013

Improvements in energy eĸciency are the most cost-eīecƟve way to deal with energy prices dispariƟes, therefore miƟgaƟng high energy costs while addressing energy security and environmental concerns. The European Union has already demonstrated how much can be done in reducing the energy intensity of manufacturing processes: its twelve largest member countries have achieved a bigger reducƟon in the relaƟve weight of energy inputs in their exports of manufactured goods than any of their external trade partners since 1995 (EC,ථ2012). In the New Policies Scenario, which assumes cauƟous implementaƟon of a raŌ of announced measures, 55% of the global economic potenƟal for improving eĸciency in industry (and two-thirds of total energy use) nonetheless remains untapped through to 2035 (see Chapter 8 | Energy and competitiveness

297

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18

Chapterථ7). The potenƟal for eĸciency gains in industry varies across sector. The most energy-intensive industries generally already use relaƟvely eĸcient technologies as they have a strong Įnancial incenƟve to save energy and boost their proĮtability. Nonetheless, there is oŌen sƟll scope for signiĮcant energy savings in energy-intensive manufacturing by replacing older faciliƟes or opƟmising processes and energy management pracƟces. There is even greater potenƟal for energy savings through the development and adopƟon of innovaƟve producƟon technologies.15 Persistently large energy price dispariƟes between regions can, in principle, drive more innovaƟon (WEF, 2013b). Box 8.6 ‫ ٲ‬Energy competitiveness and the European Union 16 Industrial and energy compeƟƟveness were key issues discussed by European Union leaders at their summit meeƟng in Brussels on 22 May 2013.16 At this meeƟng the president of the European Commission acknowledged that there is no silver bullet to boost EU’s compeƟƟveness in response to changes in global energy markets, yet indicated that there are several avenues for miƟgaƟng the negaƟve impact of persistently high energy price dispariƟes. The president set out the European Commission’s approach to a so-called ͞no regrets scenario͟, involving acƟon in Įve areas: „ CompleƟng the internal energy market. „ InvesƟng in innovaƟon and infrastructure. „ PromoƟng greater energy eĸciency. „ Using renewable sources cost-eīecƟvely. „ Diversifying energy supplies.

© OECD/IEA, 2013

But the remaining global economic potenƟal for improving energy eĸciency may not be fully realised without acƟon by governments to encourage industry to make the necessary investments, even where they ulƟmately pay for themselves. Fiscal incenƟves and Įnancing mechanisms, including tax breaks and extended payback periods, can be eīecƟve to overcome barriers to investment. SpeciĮc measures that have been shown to work well include eĸciency targets and standards, benchmarking, energy audits and energy management requirements, complemented by training, capacity-building, informaƟon provision and awareness raising campaigns. Public support for research, demonstraƟon and deployment of energy and process technologies can also deliver signiĮcant eĸciency gains.

15.ഩ Recycling could also provide a means of saving energy and effectively lowering energy costs. For example, each year the European Union disposes of Φ5.25 billion worth of recyclable goods such as paper, glass, plastics, aluminium and steel, despite having some of the highest recycling rates in the world. In theory, if all of these goods were recycled, an estimated 148 million tonnes of CO 2 emissions could be avoided annually (EC,ථ2011). 16.ഩThe European Council discussed energy and taxation in the context of the European Union’s efforts to promote growth, jobs and competitiveness. 298

World Energy Outlook 2013 | Global Energy Trends

In the World Energy Outlook 2012 Eĸcient World Scenario17 (in which eĸciency investments that are economically viable are adopted systemaƟcally due to stronger government measures) industrial energy demand growth falls to 0.8% per year on average in 2011-2035 compared with 1.2% in the New Policies Scenario. Despite an increase of around 115% in industrial sector acƟvity, energy use in the Eĸcient World Scenario increases by only 22% over the period due to energy eĸciency gains (Figureථ8.23). Most of the cumulaƟve energy savings, with respect to the New Policies Scenario, come from reduced use of electricity (37%), followed by lower use of coal (27%) and gas (18%). Emerging economies account for the majority of the cumulaƟve energy savings (China alone for 39% and India for 13%), while only 15% of savings arise in OECD countries. The potenƟal for further energy eĸciency savings is lower in OECD countries since liƩle new capacity is added over the projecƟon period and their energy intensity is in general lower than in non-OECD countries. Energy use in the Eĸcient World Scenario in 2035 is cut by 10% in pulp and paper, 8% in cement, 7% in iron and steel, and 3% in chemicals, relaƟve to the New Policies Scenario. Since signiĮcant eĸciency improvements are already part of the New Policies Scenario, addiƟonal savings from eĸciency are limited in the Eĸcient World Scenario, parƟcularly in the chemical industry where no eĸciency savings are possible for the part of the energy used as feedstock.

1

Figure 8.23 ‫& ٲ‬  RPSRXQGDYHUDJHDQQXDOFKDQJHLQLQGXVWULDOSURGXFWLRQ

9

HIÀFLHQF\DQGHQHUJ\GHPDQGE\VFHQDULR 2%

2 3 4 5 6 7 8

10

Efficiency Producon

1%

11

Energy demand

0%

12

-1%

13

-2%

NPS EWS Iron and steel

NPS EWS Chemicals

NPS EWS Cement

14

NPS EWS Pulp and paper

Notes: NegaƟve values for eĸciency represent improvements. NPS = New Policies Scenario; EWS = Eĸcient World Scenario. Source: IEA (2012b).

© OECD/IEA, 2013

Investments toward energy eĸciency in all end-use sectors in the Eĸcient World Scenario more than pay for themselves, boosƟng global GDP by an esƟmated 0.4% by 2035, as producƟon and consumpƟon of less energy-intensive goods and services free up resources 17.ഩ The additional investment in energy efficiency is in all cases economically viable. In transport, for example, the average payback period is seven years. See tKͲϮϬϭϮ for further details on the methodology used to develop the Efficient World Scenario, and results by sector, region and fuel (IEA, 2012b).

Chapter 8 | Energy and competitiveness

299

15 16 17 18

to be allocated more eĸciently elsewhere. But there are winners and losers: the energyimporƟng countries see the biggest gains, with GDP expanding by 1.1% in 2035 compared with the New Policies Scenario in OECD Europe, 1.7% in the United States, 2.1% in China and 3% in India. By contrast, GDP falls by 4.5% in Russia, as its oil and gas exports are lower. Another avenue boosƟng energy compeƟƟveness is encouraging the development of indigenous sources of energy with the potenƟal to meet domesƟc demand at lower cost. In several regions (including parts of Europe, China and LaƟn America) there is the potenƟal to replicate, at least in part, the US success in developing its unconvenƟonal gas and oil resources, but considerable uncertainty remains over the quality of the resources and the cost of producing them. Moreover, a number of technical and regulatory hurdles will need to be overcome for large-scale producƟon. What can be done to achieve this, while allaying legiƟmate public concerns about the potenƟal environmental impact, is encapsulated in the IEA’s Golden Rules (IEA, 2012c). AddiƟonally, in terms of natural gas, renegoƟaƟon of pricing terms in both exisƟng and future import contracts can be another possible avenue towards improving energy compeƟƟveness. PromoƟng gas producƟon can be compaƟble with climate goals, insofar as gas displaces more carbon-intensive coal or oil. In the longer term, even gas use will need to fall, or it will need to be used with carbon capture and storage, in order for climate goals to be met. Other low-carbon sources of energy, such as renewables and nuclear power, can contribute both to enhancing energy compeƟƟveness and achieving climate change goals. However when renewables conƟnue to receive subsidies, government support measures need to be adjusted for new capacity as technology costs and electricity prices evolve. Such eīorts will ensure that associated costs are kept to a minimum, consequently reducing the impacts on electricity prices and the burden on industries, parƟcularly for those exposed to internaƟonal compeƟƟon.

© OECD/IEA, 2013

Regardless of the make-up of energy supply, eĸcient, compeƟƟve markets are crucial to minimising the cost of energy to an economy. In many countries, market reforms aimed at liberalising energy supply and increasing compeƟƟon in wholesale and retail markets for gas and electricity are far from complete, and therefore result in an ineĸcient allocaƟon of resources and higher prices to end-users than would otherwise be the case.



World Energy Outlook 2013 | Global Energy Trends

PART B BRAZIL ENERGY OUTLOOK

PREFACE

Part B of this WEO (Chapters 9-12) conƟnues the past pracƟce of examining in depth the prospects of a country of special signiĮcance to the global energy outlook. The spotlight falls this Ɵme on Brazil. Chapter 9 surveys the situaƟon as it is today, and how historical developments have brought Brazil to this point. It also explains the analyƟcal approach for the projecƟons that follow. Chapter 10 provides a detailed analysis of Brazil’s future energy needs, projecƟng energy demand growth across all sectors and fuels, including the development of the power sector, the future role of renewables, the uƟlisaƟon of domesƟc gas supplies and the role of oil and biofuels in transport. Chapter 11 provides a detailed analysis of Brazil’s energy resources, covering the spectrum of fossil fuels, renewables and nuclear. It assesses the scale of these resources and what will be involved in their future exploitaƟon, including the potenƟal challenges and risks. The scale of necessary investment is assessed.

© OECD/IEA, 2013

Chapter 12 brings the analysis of the previous chapters together, examining the implicaƟons of Brazil’s supply and demand trends for the country itself and for the region, but also puƫng Brazilian developments in a global context. It does this by considering three dimensions of Brazilian energy: its links with economic development, energy trade and security, and the environment.

© OECD/IEA, 2013

Chapter 9 The Brazilian energy sector today Building on green foundations Highlights

x Brazil’s energy policy choices and achievements measure up well against some of the world’s most urgent energy challenges. A concerted policy eīort has meant that access to electricity is now almost universal across the country. Almost 45% of the country’s primary energy demand is met by renewable energy, making Brazil’s energy sector one of the least carbon-intensive in the world. Total primary energy demand has doubled in Brazil since 1990 on the back of strong economic growth and the emergence of a new middle class. Strong growth in electricity consumpƟon and in demand for transport fuels has led the way.

x Large hydropower plants account for around 80% of domesƟc electricity generaƟon, giving the electricity system a great deal of operaƟonal Ňexibility. ConƟnued expansion of hydropower is increasingly constrained by the remoteness and environmental sensiƟvity of a large part of the remaining resource, although 20ථGW of hydropower capacity is under construcƟon in the Amazon region.

x Reliance on other sources for power generaƟon is growing, notably natural gas, wind and bioenergy. A system of contract aucƟons provides a mechanism to bring forward investment in new generaƟon and transmission capacity, as well as to diversify the power mix.

x Biofuels, primarily sugarcane ethanol, currently meet around 15% of demand in the transport sector, where Ňex-fuel technologies account for around 90% of new passenger vehicle sales. A combinaƟon of poor harvests, rising costs, underinvestment and, since 2010, a weakened compeƟƟve posiƟon versus gasoline have held the ethanol sector back, although current market condiƟons look more promising. Biodiesel producƟon is growing and the use of bioenergy is extensive in power generaƟon and industry.

x Large oīshore oil and gas discoveries have conĮrmed Brazil’s status as one of the world’s foremost oil and gas provinces. The “pre-salt” discoveries also prompted a change in upstream regulaƟon, granƟng Petrobras – the naƟonal oil company – a strengthened role in areas deemed strategic. AŌer a Įve-year hiatus, the resumpƟon of licensing rounds in 2013 opened up new opportuniƟes to explore Brazil’s oīshore and onshore potenƟal.

© OECD/IEA, 2013

x ProducƟon from the deepwater pre-salt Įelds in the Santos basin has started, but has not yet gained suĸcient momentum to oīset declining output from mature Įelds elsewhere. Brazil’s oil output has levelled oī at just above 2ථmbͬd since 2010, and pre-salt growth will be essenƟal to re-aƩain the objecƟve of net self-suĸciency in oil and to pave the way for Brazil to become a major oil exporter. Chapter 9 | The Brazilian energy sector today

303

Introducing Brazil’s energy sector1 Brazil occupies, in many ways, an enviable posiƟon in the global energy system. Its endowment of energy resources is vast, varied and more than suĸcient to meet the country’s needs. Brazil has confronted head-on some of today’s most pressing energy challenges: almost all Brazilian households now have access to electricity and the expansion of the energy system to support a rapidly-growing economy has been achieved, to an impressive degree, through renewable energy resources. These are two of the most urgent challenges facing energy policymakers, in a world in which almost 1.3ථbillion people lack access to electricityථ(see Chapterථ2) and conƟnued reliance on fossil fuels comes at an increasingly high price, which is not yet fully reŇected in the price of fuel. Brazil’s early determinaƟon to press ahead with alternaƟves to fossil fuels was a natural choice, given the country’s large hydropower potenƟal and agricultural base, but it was also driven by concerns over energy security. DomesƟc discoveries of oil and gas were iniƟally relaƟvely modest, at least unƟl the late 1970s, and the desire to minimise reliance on imported fuels was reinforced by the oil shocks of that decade. The result of the choices made to address those challenges is that, as of 2012, around 85% of Brazil’s electricity comes from renewable sources, mainly hydropower, and in the transport sector, a stronghold of oil consumpƟon around the world, around 15% of consumpƟon is domesƟcally produced biofuels. Overall, the share of modern renewable energy in total primary energy demand in Brazil is far above the global average, making Brazil’s energy sector among the least carbon-intensive in the world (Figureථ9.1). Figure 9.1 ‫ ٲ‬Share of renewables in total primary energy demand in selected regions, 2011 50%

Tradional biomass used in the residenal sector

40%

Other renewables

30% 20% 10%

World average

OECD

NonOECD

Brazil

India

China

Russia

© OECD/IEA, 2013

Brazil is now emerging as a leading force in the oil sector. Over the last three decades, Petrobras – the naƟonal oil company – has made a series of large oīshore discoveries, iniƟally in the Campos basin, becoming a world leader in deepwater technology in the 1.ഩ This analysis has benefited greatly from discussions with Brazilian officials, industry representatives and experts, notably during a high-level WEO workshop held in Rio de Janeiro on 11 April 2013. 304

World Energy Outlook 2013 | Brazil Energy Outlook

process (Figureථ9.2). With the huge “pre-salt” Įnds in the Santos basin since 2006, Brazil’s ambiƟon in the oil sector has risen once again.2 The development of these Įelds by Petrobras and its partners will be complex and costly, but it has the potenƟal to turn Brazil into a major exporter of oil as well as a signiĮcant producer of natural gas.

1

Figure 9.2 ‫ ٲ‬Energy map of Brazil

3 km 0 200 400 600

VENEZUELA

2

4

GUYANA

FRENCH GUIANA SURINAME

COLOMBIA

5

Atlantic Ocean

6 Belo Monte

7

Fortaleza Manaus

Tucuruí

NORTH NORTH EAST

Jirau

Recifé

8

Salvador

9

BRAZIL

PERU

BOLIVIA

CENTRE WEST

10

Brasilia

11

SOUTH EAST Pacific Ocean

PARAGUAY Itaipu Curitiba

Campos Basin

12

Rio de Janeiro São Paulo Santos Basin

Atlantic Ocean

SOUTH ARGENTINA

URUGUAY

Amazon rainforest Conservation area Selected basin Gas pipeline Gas pipeline (planned) Oil pipeline Refinery Major hydropower plant

© OECD/IEA, 2013

This map is without prejudice to the status of or sovereignty over any territory, to the delimitation of international frontiers and boundaries and to the name of any territory, city or area.

2.ഩ These huge resources are called “pre-salt” because they predate the formation of a thick salt layer, which reaches up to 2ථ000 metres in places and overlays the hydrocarbons, trapping them in place (see Chapterථ11).

Chapter 9 | The Brazilian energy sector today

305

13 14 15 16

Despite Brazil’s pre-eminent posiƟon on issues of energy security, sustainability and near universal access to electricity, the challenges facing its policymakers remain considerable. Self-suĸciency in energy resources, although miƟgaƟng external risks, does not guarantee reliable supply at aīordable cost: the energy sector has already strained, on occasion, to keep up with the demands of a rapidly expanding middle class and a burgeoning economy. Although renewable resources are plenƟful, there are potenƟal limitaƟons – including social and environmental constraints – on whether their share of total energy supply can be maintained or increased. Eīorts to conserve Brazil’s biodiversity, policies on land use and water-resource management are all closely intertwined with the outlook for the energy sector. Risks to the resilience of the Brazilian power system, such as those arising from the variability of rainfall paƩerns and hydropower inŇows, could be exacerbated by a decreasing role for large storage reservoirs or by changes to the climate. And the promise of rapid growth in Brazilian oil and gas producƟon, if realised, will demand consideraƟon of new trade-oīs between economic, environmental, social and energy security objecƟves. How Brazil meets the challenges ahead will have implicaƟons not just for its own economy, but for the world at large.

DomesƟc energy trends Energy demand in Brazil has followed closely the trajectory of gross domesƟc productථ(GDP) growth over the past two decades (Figureථ9.3).3 Since 1990, energy demand has doubled, reaching nearly 270ථmillion tonnes of oil equivalent (Mtoe) in 2011. The pace of growth in both economic acƟvity and energy demand has picked up noƟceably since the turn of the century: from 2000-2011, average annual GDP growth was a full percentage point higher than in the previous decade (3.5% versus 2.5%). Oil and renewables (mainly bioenergy and hydropower) have remained dominant in the primary energy mix, the only signiĮcant change over the last two decades being the growth in demand for natural gas, which increased its share from 2% in 1990 to over 10% today.

© OECD/IEA, 2013

The aim of successive administraƟons to ensure that economic development goes hand-inhand with social inclusion has been an important determinant of energy trends. Improving access to modern energy services has been a policy priority, reŇected in targeted iniƟaƟves such as the Luz Para Todos (Light for All) programme, which was launched in 2003 with the aim of achieving universal access to electricity in Brazil by 2014. By early 2013, the programme had provided access to 14.8ථmillion people, bringing overall electriĮcaƟon rates to around 99%. The programme provides an electricity connecƟon free of charge, together with three lamps and the installaƟon of two outlets in each residence, and discounts the price for up to 220ථkilowaƩ-hours (kWh) of consumpƟon per month. This programme has been an important part of Brazil’s campaign to reduce the numbers living in extreme poverty, which fell from 17% of the populaƟon in 1990 to 6% in 2009 (UNDP, 2013). 3.ഩ The energy statistics for Brazil used here, unless otherwise specified, come from IEA databases. They may differ slightly from national statistics due to variations in methodology. We use 2011 as the base year for projections, as this is the most recent year for which a full IEA energy balance was available at the time of writing. We have incorporated 2012 data from the Brazilian government where possible. 306

World Energy Outlook 2013 | Brazil Energy Outlook

3.0

250

2.5

200

2.0

Mtoe

300

150

1.5

100

1.0

50

0.5

Trillion dollars ($2012, MER)

Figure 9.3 ‫ ٲ‬Brazil primary energy demand and GDP growth

1 Other renewables

2

Bioenergy Hydro

3

Nuclear Gas Oil

4

Coal GDP (right axis)

1995

1990

2000

2005

5 6

2010 2012

Note: MER = market exchange rate.

Broader shiŌs in Brazil’s income distribuƟon have been achieved through increased employment, improved educaƟon, income transfer policies (such as the Bolsa Familia ΀Family Allowance΁ programme) and growth in the minimum wage, which rose by 75% in real terms over 2003-2013ථ(Ministry of Finance, 2013). As a result, from 2003 to 2009 alone, around 25ථmillion people entered the middle-income group (as deĮned by the government), bringing the share of this group to above 50% for the Įrst Ɵme (Figureථ9.4). The rise of a Brazilian middle class has been a key driver of growth in energy consumpƟon. It is reŇected in the rate of passenger vehicle ownership, which has tripled since 1990. Purchases of new appliances have driven up household energy consumpƟon, with electricity use in the residenƟal and commercial sectors increasing by more than 4% per year.

Share of populaon

11%

11%

13%

14%

12%

9 10

12

High income Middle income

80%

Low income 40%

40% 47%

60%

52%

55%

13

40% 20%

49%

1999

© OECD/IEA, 2013

8

11

Figure 9.4 ‫ ٲ‬Changes in income distribution in Brazil 100%

7

49%

2003

14 40%

2006

34%

33%

2009

2011

15

Notes: This Įgure is based on the deĮniƟon of middle income used by the Government of Brazil, equaƟng to a monthly household income in 2011 of between 291 Brazilian reals (BRL) ($174) and BRLථ1ථ019 ($608). The minimum monthly income for the middle class, as deĮned by the United NaƟons, is higher at around $300. Source:ථSAE (2013).

Chapter 9 | The Brazilian energy sector today

307

16

A snapshot of energy use in 2011 shows how the diīerent fuels work their way through the Brazilian energy system (Figureථ9.5). Compared with paƩerns of energy consumpƟon elsewhere in the world, the dominance of renewables (mainly hydropower) in power generaƟon stands out, as does the relaƟvely high penetraƟon of bioenergy in industrial energy consumpƟon and transport. Fossil fuel demand in Brazil is heavily concentrated on oil products, most of which are consumed in the transport sector; natural gas (though growing fast) and coal play relaƟvely minor roles. Figure 9.5 ‫ ٲ‬Brazil domestic energy balance, 2011 (Mtoe)  

 7UDQVIRUPDWLRQ IRVVLOIXHOV



&RDO





,QGXVWU\







1DWXUDO JDV







2LO

2WKHU UHQHZDEOHV













(OHFWULFLW\ +HDW

7UDQVSRUW





 

1XFOHDU

/RVVHVDQG RZQXVH

 

%XLOGLQJV

  

%LRHQHUJ\



2WKHU

 &RQYHUVLRQORVVHV

* TransformaƟon of fossil fuels (e.g. oil reĮning) into a form that can be used in the Įnal consuming sectors. ** Includes fuel consumed in oil and gas producƟon, transformaƟon losses and own use, generaƟon lost or consumed in the process of electricity producƟon, and transmission and distribuƟon losses. *** Includes energy use in residenƟal, commercial and public buildings.

© OECD/IEA, 2013

Looking more closely at the evoluƟon of demand in the end-use sectors since 1990, energy use for transport has increased most rapidly, at an average rate of almost 4% per year (Figureථ9.6). The share of biofuels in transport demand has remained around 15%, with oil ceding a small part of its share to compressed natural gas, which has made inroads as a transport fuel in speciĮc markets, such as taxi Ňeets in São Paulo and Rio de Janeiro. With the domesƟc rail network relaƟvely under-developed, road freight has absorbed the expansion in goods traĸc generated by the growing economy. Industry is the largest of the main energy end-use sectors, its demand increasing at an average annual rate of around 3.5% from 1990 to 2011. The iron and steel industry accounts for more than one-ĮŌh of Įnal energy consumpƟon in the industrial sector, using domesƟcally produced charcoal, as well as imported coking coal. Pulp and paper processing also relies on bioenergy for a large share of its energy needs. Energy consumpƟon in the buildings sector has grown much more slowly, at just under 2% per year, in part because 308

World Energy Outlook 2013 | Brazil Energy Outlook

of a switch away from ineĸcient use of tradiƟonal biomass towards electricity. Within this sector, increased consumpƟon of electricity accounted for almost all of the growth in energy demand.

Mtoe

Figure 9.6 ‫ ٲ‬%  UD]LOÀQDOHQHUJ\FRQVXPSWLRQE\IXHOLQVHOHFWHGVHFWRUV 100

Electricity Other renewables

80

Bioenergy

1 2 3 4

Natural gas

60

5

Oil Coal

40

6

20

7 Industry Transport Buildings 1990

Industry Transport Buildings 2011

8

Power sector Since 1990, electricity supply in Brazil has more than doubled (Figureථ9.7). Hydropower remains the bedrock of the power system, accounƟng for over 70% of total installed generaƟon capacity in 2011. Depending on hydrological condiƟons, its share of total domesƟc generaƟon has typically been higher, atථ80-90%. ContribuƟons from other sources are small by comparison, but have grown quickly. Two nuclear power plants started operaƟon in 1985 and 2000. Output from bioenergy (primarily from the combusƟon of bagasse, a by-product of sugarcane processing) has increased to account for 6% of electricity generaƟon. Wind power has also been increasing its role in electricity producƟon rapidly, albeit from a very low base. The share of generaƟon from fossil-fuel power plants grew to just under 10% of the total in 2011, but this capacity was called upon much more frequently in 2012 because of concerns over hydro reservoir levels, pushing its share in total generaƟon up toථ13%. Electricity imports currently meet around 6% of total demand, with the largest amounts of imported electricity coming from the Paraguayan share of output from the bi-naƟonal Itaipu hydropower plant, which straddles the border between the two countries. This 14ථgigawaƩ (GW) facility is the world’s second-largest hydropower plant, generaƟng almost 100ථterawaƩ-hours (TWh) in 2012.

© OECD/IEA, 2013

Brazil’s hydropower plants are spread across various hydrological basins and their operaƟon is co-ordinated countrywide via a large interconnected transmission network. 4 The large hydro reservoirs give the electricity system a signiĮcant degree of Ňexibility, as the associated plants can be called upon to respond at relaƟvely short noƟce to changes in 4.ഩ This co-ordination of output from the different hydrological basins with generation from other renewable and thermal sources is done using a chain of optimisation models representing the operation of the power system. These models are also used as a planning tool for capacity and network expansion (Maceira, et al., 2002 and 2008).

Chapter 9 | The Brazilian energy sector today

309

9 10 11 12

13 14 15 16

power demand or in the availability of supply from other sources. This makes it relaƟvely straight-forward to deal with the seasonal variaƟons that are associated with bioenergy co-generaƟon, as well as the variability of other renewable sources, such as wind. This dual role of hydropower, both as a renewable source in its own right and as an enabler of other renewables, is fundamental to the Brazilian power sector’s low-carbon credenƟals and prospects. Figure 9.7 ‫ ٲ‬Brazil electricity supply by source TWh

600

Imports Other renewables

500

Oil Coal

400

Nuclear 300

Gas Bioenergy

200

Hydro

100

1990

1995

2000

2005

2011

Electricity consumpƟon has grown at just under 4% per year since 1990 – a faster rate than that of the economy as a whole. zet even though power consumpƟon has more than doubled over the last two decades, electricity use per capita is sƟll relaƟvely low by internaƟonal standards. The average Brazilian consumed around 2ථ300 kWh of electricity in 2011, 40% below the equivalent Įgure for South Africa (even though GDP per capita in the two countries is very similar) and 20% below the Įgure for China (despite Brazil’s higher per capita income). Indicators such as these help to underpin the expectaƟon of conƟnued strong growth in power demand in the years to come and the importance for Brazil, as for many emerging economies, of Ɵmely and adequate investment in both new capacity and in energy eĸciency. The risks of failing to invest adequately were highlighted in 2001-2002, when Brazil’s Ɵght supply and demand coincided with a prolonged period of lower-thanaverage rainfall and a consequent reducƟon in hydropower output, causing a major power crisis (Boxථ9.1).

© OECD/IEA, 2013

The scope for new hydropower faciliƟes is far from exhausted: Brazil has developed only one-third of its esƟmated 245ථGW of total hydro potenƟal. But while hydropower is set to retain its primary posiƟon in the power mix for decades to come, there are factors constraining its growth. Chief among these is the locaƟon of the remaining hydro potenƟal, which is concentrated in the Amazon region, far from the main centres of demand.5 Closely related issues are the environmental and social sensiƟviƟes of new projects, which Įnd 5.ഩ Around 20ථGW of hydropower capacity is under construction in the Amazon region, including the 11.2ථGW Belo Monte plant and the Jirau and Santo Antƀnio dams on the Madeira River (for another combined 6.9ථGW). 310

World Energy Outlook 2013 | Brazil Energy Outlook

expression in diĸculƟes and delays with environmental licensing and resolute opposiƟon from parts of civil society. The Brazilian authoriƟes are seeking ways to assuage public concerns and minimise social and environmental impacts, for example through the concept of “plaƞorm” hydropower development that minimises the footprint of a new project on the surrounding area (see Chapterථ11, Box 11.5). The planning process for new hydropower projects is also resulƟng in a change in the type of projects that move forward (MMEͬCEPEL, 2007). Given the topography of the Amazon region, and seeking a balance between power output, environmental and social impacts, and consideraƟons of waterresource management, most of the new hydropower plants are “run-of-river” type. These projects avoid Ňooding very extensive areas, but – as a result – have liƩle or no water storage, meaning that their power output is subject to large seasonal variaƟons. Box 9.1 ‫ ٲ‬Electricity crisis in Brazil, 2001-2002

Short-term opƟons to increase electricity generaƟon were relaƟvely limited and so the brunt of the crisis response fell on the demand side, where the government implemented a quota programme that imposed on all residenƟal, industrial and commercial consumers a monthly ceiling, set at 80% of their consumpƟon for the previous year, and penalised excess consumpƟon. This reduced electricity use by 20%, allowing Brazil to avoid the rolling blackouts that otherwise would have ensued.

© OECD/IEA, 2013

2 3 4 5 6

The roots of the 2001-2002 Brazilian power crisis can be traced to the previous decade, during which demand for electricity rose more quickly than generaƟon capacity. A farreaching reform process launched in the 1990s introduced many of the fundamentals of a compeƟƟve market, but regulatory uncertainty and weak incenƟves for distributors and large consumers to enter into long-term power supply arrangements with generators led to diĸculƟes obtaining Įnancing for new power plants. This meant that the power sector became progressively more vulnerable to the impact of adverse hydrological condiƟons. This came to a peak in the unusually dry summer of 2001: water reservoir levels in many parts of the country fell to criƟcal levels, compromising the ability to ensure reliable power supply.

The crisis had major repercussions for the Brazilian power sector. It generated new debate about how to ensure adequate investment, leading to a revised power sector model that gave the state, within a conƟnued commitment to market compeƟƟon, a more proacƟve role in planning and Įnancing new capacity. The new model obliges distributors and large consumers to cover all of their expected long-term electricity needs with long-term power purchasing agreements, providing an anchor for the system of capacity aucƟons (described below) and a more stable investment environment for new generaƟon capacity. The crisis also had a prolonged impact on demand, with higher public awareness about energy use and eĸciency meaning that total residenƟal electricity demand returned to 2000 levels only in 2005.

Chapter 9 | The Brazilian energy sector today

1

311

7 8 9 10 11 12

13 14 15 16

Energy policy in Brazil since the mid-2000s has sought to foster other sources of generaƟon. The main mechanism has been a system of contract aucƟons, in which total long-term demand from the various distribuƟon companies is matched, in a bidding process, to diīerent combinaƟons of potenƟal supply, with the most compeƟƟve bids then receiving long-term power supply contracts.6 Since 2005, 24ථaucƟons for new power generaƟon projects have been held, organised in some cases by technology, e.g.ථrenewables-only, or exclusively large hydropower (and, in some cases, only for reserve capacity). The system provides a mechanism for the authoriƟes to exert a degree of control over the evoluƟon of the power mix. Contracts for more than 500ථnew generaƟon projects have been concluded since 2005, promising to deliver around 65ථGW of capacity from a range of sources at speciĮed future dates (typically starƟng either three or Įve years aŌer the date of the aucƟon, for a period of 15 to 35 years). The aucƟons have contributed to a signiĮcant build-up of new thermal generaƟon, with the main aƩracƟon of gas-Įred power, in parƟcular, being the ease with which it can be brought online to provide back-up in case of shorƞalls elsewhere in the system. AucƟons are also used to develop the transmission and distribuƟon network. A second notable outcome of the aucƟons has been the success of wind power projects, which have competed successfully with gas-Įred power projects in some aucƟons on an (unsubsidised) cost basis. Early indicaƟons are that new wind projects are operaƟng at capacity factors in excess of 50%, high levels by internaƟonal standards. Nonetheless, there remain concerns that the intense compeƟƟon for contracts has introduced some new elements of risk, as suppliers commit to a level of long-term performance that they may be unable, in pracƟce, to deliver. In other cases, implementaƟon of new power projects or transmission lines has fallen behind schedule. As examined in more detail in Chapter 10, the contract aucƟon system has reduced, but not completely removed, uncertainty over future supply.

Bioenergy

© OECD/IEA, 2013

Brazil’s use of bioenergy is disƟncƟve, widespread and a largely successful example of government policy shaping trends in energy producƟon and use. The iniƟal spur was a naƟonal iniƟaƟve (Prſ-lcool), borne of the Įrst oil shock in the 1970s, that aimed to incenƟvise the replacement of oil as a transport fuel by ethanol produced from sugarcane. Both through the introducƟon of mandatory blending levels of ethanol with gasoline and through its sole use as a transport fuel, domesƟcally produced ethanol has regularly met between 13-21% of Brazil’s demand for road transport fuel since 1990. Overall, bioenergy accounts for more than one-quarter of primary energy demand (Boxථ9.2). 6.ഩ This system provides a contrast with the operation of power markets in many OECD countries (and the initial attempt at power sector reform in Brazil), where competition between suppliers is based on short-run marginal costs. In the Brazilian context, an emphasis on short-run marginal costs turned out to be too volatile, primarily because of the large share of hydro in the system (which is either available at very low marginal cost or, in the event of a shortfall, potentially unavailable in very large volumes), hence the preference for a system that provides more stable cash flows over time, easing project financing and investment. 312

World Energy Outlook 2013 | Brazil Energy Outlook

Box 9.2 ‫ ٲ‬Bioenergy in Brazil: more than ethanol7

1

Ethanol consumpƟon in the transport sector is the most widely known example of bioenergy use in Brazil, but it is by no means the only one. Brazil’s climate, size and the importance of its agriculture industry mean that it is well-placed to develop bioenergy, four categories of which feature in Brazil: „Firewood and charcoal. The share of wood in the Brazilian energy balance has been

declining, but sƟll accounted for some 9% of primary energy demand in 2011. Brazil has a large forestry industry and around 35% of harvested wood is transformed into charcoal, mainly for use in steel mills. Almost 40% is consumed directly in a variety of industrial and agricultural processes while the remaining 25% is used in households, primarily for cooking. „Steam turbine generaƟon systems Įred primarily by agricultural residues, such as

bagasse from sugarcane processing, and also by black liquor, a by-product from the manufacture of pulp and paper. These are oŌen co-generaƟon systems, providing on-site heat and electricity, with surplus power being sold to the grid. In 2011, 6% of the electricity used in Brazil was produced using bioenergy, half of which was sold to the market.

2 3 4 5 6 7 8

„Ethanol for the transport sector, disƟlled from sugarcane.7 This is blended with

gasoline (at a mandated level of between 18% and 25%) or used directly in Ňex-fuel or ethanol-only vehicles. Ethanol producƟon in 2012 averaged around 405ථthousand barrels per day (kbͬd). „Biodiesel for transportation. Biodiesel is produced primarily from soybean oil, with

smaller amounts from animal fats and other vegetable oils. It has grown rapidly since the launch of a state support programme in 2004 and the introduction of a blending mandate, which has risen to 5%. In 2012, 47ථkbͬd of biodiesel was produced.

© OECD/IEA, 2013

The development of bioenergy at relaƟvely low cost in Brazil has brought a range of associated economic and energy security beneĮts (especially where it displaces fossil fuel imports). The balance of environmental beneĮts is more nuanced: the CO2 released with the combusƟon of bioenergy is equivalent to the CO2 absorbed during its growth, meaning that the use of oil products during producƟon and transport is the only source of net emissions — but there is also a vigorous debate over the broader environmental impact of bioenergy, once direct and indirect changes to land use are taken into account (see Chapter 11).

7.ഩAround 60% of ethanol production in 2011 was hydrous ethanol; the rest was anhydrous. Hydrous (or wet) ethanol is produced by simple distillation and has a water content of between 4-7%. This is typically used directly as a transport fuel. The anhydrous ethanol that is blended with gasoline undergoes additional dehydration to reduce the water content to below 1%.

Chapter 9 | The Brazilian energy sector today

313

9 10 11 12

13 14 15 16

Brazil’s energy system is generally well-adapted to bioenergy use, an essenƟal component being the rapid growth over the last ten years of a “Ňex-fuel” vehicle Ňeet capable of running either on gasoline, ethanol or any mixture of the two. The popularity of a previous generaƟon of ethanol-only vehicles plummeted in the 1990s because of shorƞalls in ethanol supply and the Ňex-fuel opƟon has since become dominant in new passenger vehicle sales. Even though the technology was only introduced to the market in 2003, it already accounts for around 90% of new passenger vehicle sales and over 50% of the passenger vehicle stock is now Ňex-fuel. As drivers are able to use both gasoline and ethanol (and in some cases, also compressed natural gas), demand for these fuels has become very sensiƟve to their relaƟve prices.

Million cubic metres

Figure 9.8 ‫ ٲ‬Brazil consumption of gasoline and ethanol in road transport 50

Gasoline Anhydrous ethanol Hydrous ethanol

40

Blended gasoline

30 20 10

2005

2006

2007

2008

2009

2010

2011

2012

© OECD/IEA, 2013

Source: UNICA (2013).

In recent years, this compeƟƟon has not worked to ethanol’s advantage (Figureථ9.8). DomesƟc gasoline prices have lagged behind internaƟonal prices since the end of 2010 (as part of eīorts to diminish the impact of volaƟle internaƟonal prices and to contain inŇaƟon), boosƟng gasoline demand. This has come at a signiĮcant cost to Petrobras, which has had to import fuel at a loss to cover the gasoline balance. But factors on the supply side have also contributed to diĸcult years for the ethanol industry. Many sugarcane producers have a degree of Ňexibility in choosing whether to produce ethanol or sugar, with the proporƟons varying according to market opportuniƟes. High internaƟonal sugar prices meant a preference for sugar producƟon, which – allied to a poor sugarcane harvest in 2011 – resulted in ethanol output falling, leading the government to temporarily lower the ethanol blending mandate to 20% at the end of 2011. Market condiƟons currently look more promising, due to: a stronger harvest in 2013; more favourable economics for ethanol versus sugar producƟon; a 6.6% increase in early 2013 in the ex-reĮnery gasoline price (narrowing the gap with internaƟonal prices); a temporary reducƟon in the level of most federal taxes paid by ethanol mills and distributors; and, given expectaƟons of improved supply, a decision to restore the ethanol blending mandate toථ25%. But the industry also 314

World Energy Outlook 2013 | Brazil Energy Outlook

needs to tackle more deep-rooted issues if it is to expand in the future, notably the renewal rate of sugarcane, which is currently low, and rising input, labour and land costs that are deterring new investment.8

Oil and gas AŌer many years in which producƟon languished well behind domesƟc consumpƟon, Brazil is now recognised as a major hydrocarbons resource-holder and producer, and has become a major desƟnaƟon for internaƟonal upstream investment (following the end of Petrobras’ monopoly in 1997). This process has taken Ɵme: Brazil had to look harder for its resources than most other oil-rich countries, as the search over the years moved to ever deeper waters oīshore. But the resources are there and a series of discoveries – iniƟally concentrated in the Campos basin but then extending into the Santos basin – have conĮrmed Brazil’s status as one of the world’s foremost oil and gas provinces.

Billion boe

2 3 4 5 6 7

Figure 9.9 ‫ ٲ‬Evolution of Brazil’s proven oil and gas reserves 20

8

16

Santos

12

9

8

Campos

10

4

11

Other offshore Onshore 1950

1

1960

1970

1980

1990

2000

12

2010

Note: The lighter-shaded areas in each category are gas reserves, in billion barrels of oil equivalent.

© OECD/IEA, 2013

Sources: ANPථ(2012); Rystad Energy AS; IEA databases and analysis.

As of 2012, Brazil’s proven oil and gas reserves amount to 18.2ථbillion barrels of oil equivalent (boe) (15.3ථbillion barrels of oil and 2.9ථbillion barrels of oil equivalent – or 460ථbillion cubic metres ΀bcm΁ of natural gas) (Figureථ9.9). Over 90% of these reserves are oīshore, of which the majority is categorised as deepwater. Brazil’s promoƟon into the highest league of resource-holders started in 2006, with the discovery of what is now called the Lula Įeld in the Santos basin. This was an eye-catching Įnd for the global oil industry, not only because of its size – the largest discovery worldwide since
Chapter 9 | The Brazilian energy sector today

315

13 14 15 16

While the new pre-salt Įnds concentrated in the Santos basin underlie Brazil’s hopes of becoming a world-class oil producer and exporter, actual output is only just beginning. 9 In the meanƟme, the shallower deposits of the Campos basin in the waters oī Rio de Janeiro state remain the mainstay of Brazilian producƟon (Figureථ9.10). The challenge that is felt most strongly by Petrobras, Brazil’s dominant upstream player, has been that while the enormous Santos pre-salt projects are in their heaviest investment period, the Campos basin is facing its own problems, including declining producƟon from some of the more mature Įelds. This explains the ŇaƩening of Brazil’s producƟon curve since 2010 and the slight fall in output in 2012. Alongside strong domesƟc oil demand growth, this also explains why re-gaining net self-suĸciency in oil, a goal eīecƟvely reached for Įve years from 2006, is now dependent on the build-up of producƟon from the pre-salt Įelds. Figure 9.10 ‫ ٲ‬Brazil oil production and domestic demand 2.5

mb/d

Santos Campos

2.0

Other offshore Onshore

1.5

Domesc demand 1.0 0.5

1990

1995

2000

2005

2010 2012

© OECD/IEA, 2013

Drilling into and producing from pre-salt reservoirs requires Petrobras and its partners to overcome several formidable technical and environmental challenges. The ocean at the locaƟon of the pre-salt Įelds (200-300ථkilometres oīshore) is oŌen more than 2ථ000ථmetres deep (a depth oŌen classiĮed as ultra-deepwater) and the well needs to extend through another 5ථ000ථmetres of rock, including up to 2ථ000ථmetres of salt layers that provide a high-pressure, corrosive environment exerƟng considerable stress on the wellbore. ProducƟon above 300ථkbͬd from pre-salt Įelds (as of mid-2013) indicates that key technical and geological challenges are being overcome. But scaling up producƟon remains a huge task, necessitaƟng a step-change in investment levels over the coming years. The requirement that a large part of the construcƟon and supplies for all the wells, faciliƟes and infrastructure be sourced locally within Brazil is sƟmulaƟng local industrial development, but adds potenƟally important strain to the supply chain in the coming years. The oil in the pre-salt Įelds is of sweet, light quality and contains signiĮcant amounts of dissolved gas (including CO2), raising expectaƟons in some quarters that Brazil will become 9.ഩ Pre-salt fields have also been found in the Campos basin, although the accumulations are smaller than those discovered in Santos. 316

World Energy Outlook 2013 | Brazil Energy Outlook

a major producer of natural gas. In 2012, domesƟc producƟon of natural gas reached 18ථbcm (net of reinjecƟon and Ňaring) and the remaining share of domesƟc consumpƟon was met either by pipeline imports from Bolivia, or, to a lesser extent, by imported liqueĮed natural gas (LNG). Future producƟon growth could come in part from onshore, where there are signs of renewed interest in exploring and developing Brazil’s gas potenƟal, including its unconvenƟonal gas resources. But the greatest uncertainty surrounds the volumes of associated gas that may become available from the deep oīshore, with a criƟcal and as yet unknown variable being the volumes of gas that may be required for reinjecƟon to maintain reservoir pressureථ(see Chapter 11).

1

Box 9.3 ‫ ٲ‬Brazil’s upstream regulatory framework10

5

There are now three systems governing upstream hydrocarbon acƟvity in Brazil: the concessionary system; a special producƟon-sharing regime for new developments in the main pre-salt area (which could be extended in the future to other areas idenƟĮed as being of strategic importance); and a system which grants deposits to Petrobras under a “transfer of rights” programme from the government (also known as the Onerous Assignment Law).

2 3 4

6 7 8

„Under the exisƟng concessionary system, any company can parƟcipate in the

various licensing rounds and there is no obligatory state parƟcipaƟon in projects (although, in pracƟce, Petrobras has remained the dominant player, with interests in many of the most prospecƟve areas). AŌer payment of royalƟes and taxes, the oil produced belongs to the concession-holder. „For any new blocks opened up in the designated area of pre-salt potenƟal (see map

in Chapterථ11, Figureථ11.7), Petrobras has to be the operator and hold a minimum 30% interest. The concession-based system is replaced by a producƟon-sharing mechanism, with the share of proĮt oil10 oīered to the state the key parameter in the contract award. „In some pre-salt areas that have not been oīered for external investment, the

government has capitalised Petrobras with a direct right to develop up to 5ථbillion barrels of reserves. The reserves involved are commonly known as “transfer of rights”.

© OECD/IEA, 2013

As part of a strategy to encourage development of the Brazilian oil and gas service sector, the requirement to source a certain share of goods and services from within Brazil has become increasingly important. Local content requirements have been sƟpulated in each licensing round and have been raised over Ɵme. Interested companies in many cases commiƩed to a level of local content in excess of the basic requirement in order to increase their chances in the assessment of bids.

10.ഩProfit oil is the amount of production, after deducting production allocated to costs and expenses (“cost oil”), which is divided between the parties and the host government under the production-sharing contract.

Chapter 9 | The Brazilian energy sector today

317

9 10 11 12

13 14 15 16

The Santos basin discoveries have not only changed the outlook for Brazilian producƟon, but also the overall approach to the exploitaƟon of Brazil’s upstream resources. In the second-half of the 1990s, Brazil’s regulatory system underwent a major overhaul, with the decision to end Petrobras’ monopoly in the oil and gas sectors and to open the upstream to internaƟonal investment. A series of ten licensing rounds from 1999 to 2008 saw acreage awarded to 78ථcompanies, Brazilian and internaƟonal. But the scale and success rate of pre-salt discoveries led the Brazilian authoriƟes to conclude that, for these resources, the concession-based system had to change. As a result, in a designated geographical area that covers the parts of the Campos and Santos basins with pre-salt potenƟal, the pendulum has swung back towards a guaranteed role for Petrobras and a diīerent system of resource management, involving a higher government take (Boxථ9.3). AŌer a Įve-year gap, a further concession-based licensing round — the eleventh — saw 142ථblocks awarded (87 onshore, 55 oīshore) in May 2013 and a twelŌh round, focusing on onshore gas, is scheduled for November. A Įrst licensing round under the producƟon-sharing system, for the right to develop the huge Libra pre-salt prospect in the Santos basin, was held in October 2013.

Energy-related CO2ĞŵŝƐƐŝŽŶƐĂŶĚĞŶĞƌŐLJĞĸĐŝĞŶĐLJ Brazil’s high share of low-carbon energy in its energy mix yields a low Įgure for energyrelated carbon-dioxide (CO2) emissions, 409ථmillion tonnesථ(Mt) in 2011. This is onequarter of the energy-related emissions in Russia, even though the Brazilian economy is one-ĮŌh larger. Brazil is unusual in that historically its energy sector has not been the largest source of naƟonal greenhouse-gas emissions (Figureථ9.11). In 2005, the energy sector was responsible for just 16% of total greenhouse-gas emissions, with the largest contribuƟon coming from land use, land-use change and forestry (LULUCF). Since 2005, Brazil has embarked on a large-scale campaign to slow deforestaƟon. As greenhouse-gas emissions from LULUCF have declined, the share of the energy sector in total emissions has doubled, to 32% in 2010, second only to emissions from agriculture (35%). Figure 9.11 ‫ ٲ‬Brazil greenhouse-gas emissions by source Waste

2 500

Land use, land-use change and forestry

Mt of CO2-eq

3 000

Agriculture

2 000

Industrial processes 1 500

Energy

1 000

© OECD/IEA, 2013

500

1990

1995

2000

2005

2010

Source: Ministry of Science and Technology (2013). 318

World Energy Outlook 2013 | Brazil Energy Outlook

Brazil’s energy intensity (primary energy demand per unit of GDP), an indicator that is someƟmes used as a proxy for the overall eĸciency of energy use (see Chapterථ7, Box 7.2), is comparable to the OECD average. Based on data for 2011, it takes 0.11ථtonnes of oil equivalent (toe) on average to create $1ථ000 of GDP (at market exchange rates) in Brazil, compared with 0.12ථtoe in OECD countries. By contrast, the global average was 0.19ථtoe and the average for the other BRICS11 (excluding Brazil) was 0.36ථtoe. Among the factors that contribute to Brazil’s low energy intensity, two stand out: the relaƟvely small amount of energy used for heaƟng (and cooling) and the large share of hydropower in the energy system. There are no, or only very limited, conversion losses from hydropower so it is a highly eĸcient form of power generaƟon compared with electricity generated from fossil fuels. While the absolute indicators for energy intensity in Brazil are impressive, the trends are less so. Brazil’s energy intensity has remained at roughly the same level for the last two decades, while there has been a slow but steady improvement in many other countries and regions. As a result, Brazil is moving steadily closer to global and regional averages (Figureථ9.12). Figure 9.12 ‫ ٲ‬Energy intensity of GDP in Brazil as a share of selected regional and global averages 100%

1 2 3 4 5 6 7 8

Share of OECD average

9

80% Share of world average

60% 40%

Share of BRICS average

10 11

20%

12 1990

1995

2000

2005

2011

© OECD/IEA, 2013

ZĞŐŝŽŶĂůĂŶĚŐůŽďĂůŝŶƚĞƌĂĐƟŽŶƐ

13

Brazil’s large and growing domesƟc market limits to some degree its reliance on internaƟonal trade and its exposure to internaƟonal markets. Total trade accounts for only around one-quarter of GDP, around half the average in the other BRICS. Nonetheless, by virtue of its size and economic weight, Brazil remains crucial to the well-being of its region and to the prospects for its conƟnued integraƟon. The Brazilian economy is Įve Ɵmes the size of the next largest in South America (ArgenƟna) and is an important driver of regional trade. Around 10% of the region’s exports in 2011 were desƟned for the 11.ഩ Brazil, Russia, India, China and South Africa.

Chapter 9 | The Brazilian energy sector today

319

14 15 16

Brazilian market. Energy plays an important role in this trade relaƟonship, with ArgenƟna, Uruguay, Venezuela, Colombia and Peru exporƟng petroleum products and coal to Brazil. ParƟcularly important for Brazil are its agreement with Paraguay to purchase electricity from the Itaipu hydropower plant that is jointly owned by the two countries and the 1999 gas trade agreement with Bolivia that provides Brazil with some 10ථbcm of natural gas per year. HighlighƟng Brazil’s value to its neighbours, gas exports to Brazil from Bolivia were worth more than $3ථbillion in 2012, over 30% of total Bolivian export earnings. The compleƟon of the Bolivia-Brazil natural gas pipeline in 1999 helped to push more gas into the Brazilian energy mix and the share of imported gas in total demand has risen steadily since then. The Bolivia pipeline remains the main source of imports but, since 2009, these have been supplemented by LNG imported via two regasiĮcaƟon terminals located in the northeast and southeast of Brazil. And despite the rise in Brazil’s oil output over the last ten years, a shortage of reĮning capacity has meant a conƟnued reliance on imported diesel, naphtha and liqueĮed petroleum gas (LPG) and, since 2011, a switch from net exports to net imports of gasoline (Figureථ9.13).12  UD]LOVHOIVXIÀFLHQF\LQQDWXUDOJDVDQGVHOHFWHGRLOSURGXFWV Figure 9.13 ‫ ٲ‬% 100%

Diesel Gasoline

80%

Natural gas

60% 40% 20%

2000

2005

2012

Source: EPE (2010 and 2013a); IEA databases.

© OECD/IEA, 2013

Brazil has a diverse set of internaƟonal relaƟonships that bear on the energy sector, among which the growing partnership with China stands out. Over 2000-2011, total bilateral trade between China and Brazil increased 33-fold, reaching $77ථbillion, making China Brazil’s most important trade partner. China’s demand has even shiŌed the overall balance of Brazil’s exports away from manufactured and semi-manufactured goods towards primary commodiƟes (the laƩer’s share in Brazilian exports has risen to 50%), leading to concerns 12.ഩ The nameplate capacity of Brazil’s refineries is around 2ථmbͬd, with several additional projects underway or planned. Although current oil output is close to this level, Brazil currently produces mostly heavier crudes, not all of which can be processed in the domestic refining system. Some light sweet crude from West Africa is imported to boost output of the lighter products. 320

World Energy Outlook 2013 | Brazil Energy Outlook

about potenƟal “de-industrialisaƟon” and increased exposure to internaƟonal commodity prices. Chinese companies have also become important investors in Brazil, both in the power sector, and in oil and gas.13 In all, between 2005 and 2012, China invested $18.2ථbillion in Brazil’s energy sector, accounƟng for 70% of total Chinese investment in the country, with deepwater experƟse and technology a parƟcular point of aƩracƟon. A second set of important external relaƟonships is with African countries, where Brazil has generated substanƟal interest as a model for sustainable growth in the developing world. Brazil has made considerable eīorts to improve Ɵes with Africa over the last decade, doubling its number of embassies across the conƟnent. Although Brazil’s trade with Africa accounted for only around 5% of Brazil’s total trade in 2012, it has increased four-fold since 2002. Energy linkages, in parƟcular, are expanding rapidly, based on a desire to emulate Brazilian successes in biofuels producƟon, deepwater drilling and mining. Examples of recent Brazilian companies’ involvement in Africa include Odebrecht’s joint venture with Sonangol and Damer Industria to produce ethanol; Petrobras’ deepwater acƟviƟes in Angola; and Vale’s recent opening of a coal mine in Mozambique.14

The projecƟons for Brazil, developed in subsequent chapters, follow the overall analyƟcal approach that is taken elsewhere in this Outlook. The primary focus for analysis is the New Policies Scenario, which takes into account both exisƟng policies and modest realisaƟon of Brazil’s policy intenƟons. Around the projecƟons for the New Policies Scenario, we include some case studies assessing speciĮc variaƟons in policy or circumstance, notably the potenƟal impact of stronger acƟon on energy eĸciency and the possibility that the contribuƟon from the hydropower sector might be held back by regulatory or climaƟc factors. In relaƟon to oil, we consider a case in which Brazil’s producƟon increases more quickly than we project in the New Policies Scenario. The projecƟons for Brazil are naturally subject to a range of uncertainƟes relaƟng to economic development, demographics, prices and policies. Baseline factual informaƟon for the analysis, though, is commendably clear, thanks to the availability and quality of the energy data collected by the Brazilian government, which helps to inform policymaking across a range of insƟtuƟons (Figureථ9.14). Diīerences in methodology mean that the IEA energy data for Brazil, used in this report, in some respects vary from the data published by the Brazilian authoriƟes.

© OECD/IEA, 2013

2 3 4 5 6 7 8

ProjecƟng future developments

13.ഩ Investments include State Grid Corporation’s acquisition of seven Brazilian power companies, Sinochem’s acquisition of Statoil’s stake in the Peregrino field, and Sinopec’s acquisition of a 30% stake in Galp Energia’s Brazilian assets and a 40% stake in Repsol zPF (the second-largest holder of exploratory rights after Petrobras in the Santos, Campos and Espirito Santo basins) 14.ഩ Portuguese-speaking parts of Africa are natural target markets for Brazilian energy and engineering companies. African members of the Community of Portuguese Language Countries (CPLP) include Angola, Mozambique, Cape Verde, Guinea-Bissau, São Tomé and Prşncipe and Equatorial Guinea (associate observer).

Chapter 9 | The Brazilian energy sector today

1

321

9 10 11 12

13 14 15 16

Figure 9.14 ‫ ٲ‬Brazil’s energy policy and regulatory institutions Other ministries represented in the CNPE: Planning, Budget and Administration; Treasury; Environment; Development, Industry and Foreign Trade; National Integration; Agriculture, Livestock and Supply

National Energy Policy Council (CNPE) A high-level multi-sectoral advisory panel for energy policy, chaired by the Minister of Mines and Energy

Electric Sector Monitoring Committee (CMSE) Monitors security of power supply

Ministry of Mines and Energy (MME) Elaborates policies for energy and mineral resources following CNPE guidelines

National Agency for Petroleum, Natural Gas and Biofuels (ANP) Implement policies, regulates and supervises the production and distribution of fuels in Brazil

Energy Research Office (EPE) Produces Brazil's energy balance and longterm energy plans

National Electric Energy Agency (ANEEL) Implements policies, regulates and supervises the power sector

Chamber for the Commercialisation of Electrical Energy (CCEE) Wholesale market operation (accounting and clearing)

National Electrical System Operator (ONS) Co-ordinates and controls power generation and transmission

The building blocks ĐŽŶŽŵŝĐĂŶĚƉŽƉƵůĂƟŽŶŐƌŽǁƚŚ

© OECD/IEA, 2013

With GDP of $2.3ථtrillion in 2011 (year-2012 dollars in purchasing power parity terms) Brazil’s economy is among the ten largest in the world. It has grown by nearly 50% in real terms since 2000, with growth averaging 3.5% per year to 2011, and conƟnued to perform relaƟvely well in the period immediately following the global economic crisis. However, Brazil has not been immune to the crisis and has taken policy acƟon in response to strong capital inŇows and currency appreciaƟon (which have since moderated), above-target inŇaƟon and weak global demand. GDP growth of 7.5% in 2010 and 2.7% in 2011 was followed by lower than expected growth of 0.9% in 2012. DomesƟc demand has been central to conƟnued economic expansion, underpinned by moderate credit expansion, job creaƟon and income growth (Central Bank of Brazil, 2013a). The unemployment rate has generally been on a declining trend, with the naƟonal rate standing at 5.6% in July 2013 and even lower rates in some important energy producing areas, Ğ͘Ő͘ 4.7% in Rio de Janeiro Stateථ(Central Bank of Brazil, 2013b). The services sector has been performing strongly and now accounts for more than two-thirds of Brazil’s GDP. Industry makes up 27% of the economy and Brazil’s is the second-largest industrial sector in the Americas. While Brazil is a heavyweight in terms of natural resources and has one of the world’s largest oil and gas companies in Petrobras, its economy is not heavily dependent on the energy sector.

322

World Energy Outlook 2013 | Brazil Energy Outlook

Among the factors that may limit Brazil’s growth prospects in the medium to long term are the state of the country’s infrastructure, the availability of a suĸciently skilled workforce and the set of constraints arising from a generally complex business and regulatory environment, oŌen known collecƟvely as the ĐƵƐƚŽ ƌĂƐŝů (Brazil cost). The government is seeking to address these issues. On infrastructure, for example, the 2011-2014 Growth AcceleraƟon Programme allocated over $57ථbillion to improving transport infrastructure and a further $255ථbillion to a wide-ranging energy plan that covers power generaƟon and transmission, oil and gas exploraƟon, producƟon and research. The economic raƟonale for these investments is clear, but the scale and pace at which Brazil is seeking to upgrade its infrastructure carries with it the risk of delay. In line with the assumpƟons used elsewhere in this Outlook, the medium-term GDP growth outlook for Brazil is based on the projecƟons of the InternaƟonal Monetary Fund (see Chapter 1). The average growth rate for 2011-2020 is held back by the low growth recorded in 2012. However, the longer-term GDP assumpƟons move slightly higher than in WEO-2012, reŇecƟng the view that the prospects for growth and producƟvity gains remain robust even if, in some sectors, these will be deferred to a later period. We assume Brazil’s GDP grows by an average of 3.7% per year over the period 2011-2035, slightly higher than the global average (Tableථ9.1). GDP per capita increases by 3% per year on average, rising from around the world average in 2011 to 13% above it in 2035. Table 9.1 ‫ ٲ‬GDP and population indicators and assumptions GDP

1 2 3 4 5 6 7 8 9

PopulaƟon 20112020* (%)

20212035* (%)

20112035* (%)

10

197

0.8%

0.5%

0.6%

3.1%

6ථ960

1.1%

0.8%

0.9%

11

5.4%

2ථ785

0.8%

0.4%

0.5%

2011 ($ billion)

20112020* (%)

20212035* (%)

20112035* (%)

Brazil

2ථ375

4.1%

3.6%

3.7%

World

69ථ937

4.0%

2.9%

BRICS**

11ථ976

8.0%

4.4%

2011 (million)

12

© OECD/IEA, 2013

* Compound average annual growth rate. ** BRICS excluding Brazil. Notes: PopulaƟon esƟmates and projecƟons in the WEO are based on those of the United NaƟons PopulaƟon Division. The latest UN esƟmate of Brazil’s populaƟon in 2011 diīers from Brazil’s oĸcial dataථ(193ථmillion).

Brazil is the ĮŌh-most populous country in the world, with around 195ථmillion people in 2011. Since 2000, the populaƟon has grown by just over 1% per year on average, but the rate of increase has been slowing gradually. About 166ථmillion Brazilians live in urban areas: this is a relaƟvely large share for the country’s stage of development, explained by historically high rates of populaƟon growth in towns and ciƟes, rural to urban migraƟon and the urbanisaƟon of former rural areas. The vast size of the country results in a relaƟvely low populaƟon density (23ථpeople per square kilometre) but, in reality, a large share of the populaƟon is concentrated in urban areas along the coast. More than 40% of the populaƟon lives in the southeast of the country and more than one-quarter lives in the northeastථ(IBGE, 2011).

Chapter 9 | The Brazilian energy sector today

323

13 14 15 16

PopulaƟon growth is an important driver of energy use, directly through its impact on the size and composiƟon of energy demand and indirectly through its eīect on economic growth and development. The populaƟon assumpƟons are based on the medium variant of the latest United NaƟons projecƟons (see Chapterථ1), which see Brazil’s populaƟon growing to 226ථmillion in 2035, increasing by 0.6% per year, on average, compared with world populaƟon growth of 0.9% per year. The rate of populaƟon growth slows over Ɵme, from around 0.8% per year, on average, before 2020 to around 0.5% per year aŌerwards. The urbanisaƟon rate in Brazil conƟnues to increase, going from 85% in 2011 to 89% by 2035 and the median age also rises from 29 years in 2010 (in line with the world average at that Ɵme) to 39 years in 2035. The proporƟon of the populaƟon that is of working age (15-64ථyears) is projected to peak near 70% around 2020-2025, but the absolute size of the working age populaƟon conƟnues to grow unƟl near the end of the Outlook period.

ŶĞƌŐLJƉƌŝĐĞƐ Energy prices in the tŽƌůĚ ŶĞƌŐLJ KƵƚůŽŽŬ are determined as a product of the World Energy Model, rather than imposed as assumpƟons (see Chapterථ1). But, in this analysis, parƟcular account needs to be taken of the condiƟons in Brazil. As a large and rapidly growing economy, Brazil has had to balance signiĮcant growth in demand with ensuring economic condiƟons that bring forth the necessary increase in supply. In line with a gradual process of liberalisaƟon that started in the 1990s, most energy prices in Brazil move either directly or indirectly in response to market signals, but there have nonetheless been regular instances of government intervenƟon to keep prices in check, moƟvated by public policy objecƟves, such as the desire to maintain industrial compeƟƟveness or eīorts to keep inŇaƟon down. If such intervenƟons on energy prices were to be reinforced over the longer term, there would be a material impact on the evoluƟon of the Brazilian energy system, encouraging more rapid growth in demand while limiƟng the incenƟve to invest in supply and energy eĸciency.

© OECD/IEA, 2013

For oil products, the core assumpƟon is that price movements in Brazil will be enƟrely aligned with internaƟonal market dynamics. This implies, for example, an end to the current under-pricing of gasoline. A parƟal excepƟon to this assumpƟon is policy in relaƟon to LPG, which is priced to encourage the use of this fuel in the residenƟal sector. High oil product prices imply a boost to the compeƟƟve posiƟon of ethanol in the transport sector and of natural gas in the industrial, commercial and residenƟal sectors, as long as the cost pressures on ethanol and natural gas are manageable. Future natural gas prices in Brazil are subject to a wide range of uncertainƟes related to internaƟonal market condiƟons, the evoluƟon of the Brazilian supply and demand balance and the possible changes to the current market structure, which has Petrobras in a dominant posiƟon in all areas of the gas supply chain. As of 2012, prices on the domesƟc market are typically pegged at 90% of the cost of fuel oil (for the same energy

324

World Energy Outlook 2013 | Brazil Energy Outlook

content), which meant $8-10 per million BriƟsh thermal units (MBtu) in 2012.15 These prices have risen substanƟally since the mid-2000s, helping Petrobras to cover the costs of new gas producƟon and related infrastructure. Over the projecƟon period, we assume that domesƟcally produced gas, which becomes available in much larger quanƟƟes in the projecƟons, will conƟnue to be priced in a way that allows for its absorpƟon on the domesƟc market, notably for power generaƟon and industrial uses. As for imported sources of gas, the average price of LNG imported to Brazil in 2012 was above $12ͬMBtu (this has risen further in the Įrst half of 2013), while pipeline imports from Bolivia cost around $10-11ͬMBtu. In the New Policies Scenario, the natural gas import price, reŇecƟng the average cost of gas imports, remains within the range of $11-13ͬMBtu over the period to 2035 (in year-2012 dollars). As to end-user prices, these vary widely in diīerent parts of the country, but the current average price paid by industry is around $17ͬMBtu; this is above the OECD average and four Ɵmes more than the price paid by industrial consumers in North America in 2012. We discuss the market and regulatory factors that could inŇuence the evoluƟon of domesƟc gas prices in subsequent chapters. Electricity prices in Brazil have risen in recent years and, by 2012, the average price paid by industry had reached $178 per megawaƩ-hour (MWh). This was at the upper range of prices paid in OECD countries and well above the level of other BRICS. The average prices paid by residenƟal users, at around $237ͬMWh, were at the middle of an OECD range, but were high by comparison with other emerging economies. Concern about the impact of these higher prices on the Brazilian economy led the government to renew around 20ථlarge power generaƟon concessions that were due to expire between 2015 and 2017, in exchange for reduced power costs. Together with a reducƟon in some taxes, this allowed for a lower power price for industry of up to 28% and 16% for households. This new (2013) price structure is taken as the baseline for the evoluƟon of electricity prices.

Policies

© OECD/IEA, 2013

The breadth and quality of Brazil’s endowment of energy resources have allowed policymakers to chart a disƟncƟve path as they pursue the tradiƟonal trinity of energy policy concerns: security, aīordability and sustainability. In the Brazilian case, large-scale deployment of indigenous bioenergy and hydropower has enabled the country to limit its reliance on imported fossil fuels, beneĮt from relaƟvely low-cost energy (at least unƟl quite recently), expand access to modern energy services and become a world leader in low-carbon energy development. This virtuous circle has served Brazil well and remains at the heart of Brazilian energy policymaking; but there are signs that the tradiƟonal alignment of energy goals is shiŌing. The energy needs of the economy are expanding fast, bringing new sources of energy (both renewable and non-renewable) into the energy mix. Large oil and gas discoveries mean the

15.ഩ Excluding local distribution company tariffs and taxes.

Chapter 9 | The Brazilian energy sector today

325

1 2 3 4 5 6 7 8 9 10 11 12

13 14 15 16

Table 9.2 ‫ ٲ‬Main policy assumptions for Brazil in the New Policies Scenario Cross-cuƫng: climate and emissions „ 36% reducƟon in greenhouse-gas emissions compared with business-as-usual by 2020 (the lower end of the range speciĮed in the NaƟonal Climate Change Policy). „ Emissions trading scheme for Rio de Janeiro state from 2014, incorporaƟng major sectors for industrial emissions. Cross-cuƫng: energy eĸciency „ Further implementaƟon of the measures in the NaƟonal Plan for Energy Eĸciency, including an extension and Ɵghtening of the Brazilian Labelling Programme (PBE), the NaƟonal Programme for Energy ConservaƟon (PROCEL), the NaƟonal Programme for RaƟonal Use of Oil Products and Natural Gas (CONPET), and an extension of the scope of eĸciency standards for equipment and machinery. „ A conƟnuaƟon of the Energy Eĸciency Programme (PEE) under which uƟliƟes must spend at least 0.5% of their operaƟng revenues on energy eĸciency. Cross-cuƫng: other „ Policies for increasing natural gas supply, restricƟng gas Ňaring and expanding gas pipeline infrastructure (Ten-zear Plan for Expansion of Gas Pipelines – PEMAT). Power sector „ Targeted aucƟons to maintain a strong renewables-based share in the power sector. „ ReducƟon of non-technical losses in the power sector. „ Special funding condiƟons to promote network metering; support (through the Inova Empresa programme) for smart grid technology and its deployment. Transport „ Ethanol blending mandate at the upper limit of an 18-25% range, plus Ňex-fuel passenger lightduty vehicle (PLDV) Ňeet consuming ethanol. „ Voluntary fuel eĸciency labelling for PLDVs; support (through the Inovar-Auto programme) for vehicle energy eĸciency and hybrid technologies; vehicle polluƟon control measures through the PROCONVE programme. „ Biodiesel blending mandate of 5% (at present), with a gradual rise in the mandated share over the projecƟon period. „ Concessions to improve port, road, rail and air infrastructure, as per the Accelerated Growth Programme 2011-2014. „ Long-term plan for freight transport (PNLT), developed by the Ministry of Transport. „ NaƟonal urban mobility plan (PNMU), developed by the Ministry of CiƟes. Industry „ Local content requirements in the oil, gas and power sectors. „ Enhanced eĸciency measures in line with the NaƟonal Energy Eĸciency Plan. „ Funding from the NaƟonal Climate Fund and from the Brazilian Development Bank’s PROESCO programme for energy eĸciency projects. Buildings „ Enhanced eĸciency measures in line with the NaƟonal Energy Eĸciency Plan.

© OECD/IEA, 2013

„ Measures to encourage the deployment of end-use solar photovoltaic applicaƟons. „ Network metering.

326

World Energy Outlook 2013 | Brazil Energy Outlook

concerns over import dependence are no longer as prominent as they once were: the central raƟonale for conƟnued large-scale renewable deployment is, instead, to sustain an important area of naƟonal industrial experƟse and to miƟgate the rise in CO2 emissions. The social and environmental dimensions of energy development, including water and landuse issues, are gaining in importance, parƟcularly for projects that have direct or indirect impacts on the Amazon region. The links between energy and economic development are also being recast. Policymakers are increasingly concerned about the impact of sharply higher energy prices on the naƟonal economy. They are also keen to ensure, via local content requirements, that the investment in new energy Įelds, notably the pre-salt hydrocarbon resources but also non-hydro renewables, brings direct local economic beneĮts. Against this increasingly complex backdrop, the policy commitments, announcements and intenƟons of the Brazilian authoriƟes – and the extent to which these are implemented successfully – are of fundamental importance in shaping the outlook for the energy sector. Brazil has a well-developed insƟtuƟonal and policy framework for the sector (Figureථ9.14), as well as a system of detailed operaƟonal planning for its expansion, based on Brazil’s expected energy needs as well as consideraƟon of social and environmental aspects. The resulƟng long-term and ten-year expansion plans are key points of reference for the energy policy outlook in Brazil. The long-term expansion plan, oŌen referred to as PNE-2030 (EPE, 2007) is in the process of being updated (and its horizon extended from 2030 to 2050) but, even in its present form, provides some important guidance on long-term policy objecƟves. The ten-year expansion plan, which is updated every year and currently looks out to 2021 (EPE, 2013b) provides a detailed sector-by-sector analysis of the anƟcipated development of the energy system and builds on engineering and environmental studies of speciĮc projects scheduled for implementaƟon (for example, via the system of aucƟons for new generaƟon and transmission capacity). The NaƟonal Policy on Climate Change (Interministerial CommiƩee on Climate Change, 2008), adopted in 2009, idenƟĮes speciĮc acƟons and measures that can miƟgate greenhouse-gas emissions, including a speciĮc target for emissions to 2020. In addiƟon, there are documents dealing with speciĮc policy areas, notably the NaƟonal AcƟon Plan for Energy Eĸciencyථ(PNEF) (MME, 2010), and policy documents related to other issues – water, biodiversity, land use, deforestaƟon, conservaƟon units, societal consent, etc. – that require integraƟon into coherent energy policymaking. Some key measures from these and other policy documents that are considered in the New Policies Scenario are listed in Tableථ9.2.

1 2 3 4 5 6 7 8 9 10 11 12

13 14

© OECD/IEA, 2013

15 16

Chapter 9 | The Brazilian energy sector today

327

© OECD/IEA, 2013

Chapter 10 Prospects for Brazil’s domestic energy consumption *RLQJZLWKWKHÁRZ Highlights

x Brazil’s primary energy demand rises by 80% in the New Policies Scenario to reach 480ථMtoe in 2035, spurring and accompanying steady growth in economic output. Growth in electricity demand is parƟcularly strong, doubling to reach 940ථTWh. Brazil achieves its goal of providing universal energy access early in the projecƟon period.

x Despite increasing availability of domesƟc oil and gas, renewable sources of energy retain their disƟncƟve posiƟon in Brazil’s primary energy mix, their share remaining Įrm at 43% in 2035. Among the other fuels, the share of oil in energy demand declines from 41% to 34%, while the share of natural gas increases from 9% to 16%. The rise in gas use hinges criƟcally on the establishment of a long-term framework for the sector that is aƩracƟve to new suppliers and consumers.

x Hydropower capacity increases by almost 70ථGW in the New Policies Scenario, but this expansion depends on suĸcient social consent for the licensing and implementaƟon of new projects. If hydropower expands more slowly, then it is likely that all other technologies would be called upon to Įll the gap, meaning accelerated growth for other renewables but also the likelihood of addiƟonal nuclear and fossil-fuel capacity, the laƩer pushing up CO2 emissions.

x Most of the anƟcipated growth in hydropower capacity is expected to come from runof-river projects, which increases the conƟngency of power output on natural and seasonal variaƟons. The Ňexibility aīorded by exisƟng reservoirs and the seasonal supply paƩerns of wind and bioenergy help to balance this, but gas-Įred capacity remains prized as a reliable complementary source of power. Overall, the power sector needs more than $555 billion in investment through to 2035 ($24 billion per year on average), of which 45% is on transmission and distribuƟon.

x Electricity demand growth is strongest in residenƟal and commercial buildings through to 2035, highlighƟng the role of appliance standards and other eĸciency policies in relieving potenƟal stress on the power sector. Industrial energy use rises by 2.5% per year, but the relaƟvely high cost of electricity and natural gas in Brazil is a factor holding back the growth of energy-intensive industry in our projecƟons.

© OECD/IEA, 2013

x Biofuels account for nearly one-third of the energy used in road transport by 2035. The energy performance of the transport sector is enhanced by new fuel eĸciency policies for passenger vehicles and a shiŌ away from today’s heavy reliance on road transport for freight. With these factors combining to slow the growth in demand for oil products, parƟcularly for gasoline, new reĮnery construcƟon allows Brazil to meet all of its domesƟc oil product needs by around 2020. Chapter 10 | Prospects for Brazil’s domestic energy consumption

329

DomesƟc energy consumpƟon trends Rising energy consumpƟon in Brazil is set to accompany and spur growth in naƟonal income over the Outlook period. In our projecƟons, primary energy use increases to 2035 by between 56% and 88%, depending on the scenarioථ(Figureථ10.1). In all of our scenarios, the percentage increase in energy demand is bigger than the equivalent projecƟon for China and second only to that for India among the BRICS (Brazil, Russia, India, China and South Africa). The eventual trajectory will depend on a range of factors, parƟcularly the rate of gross domesƟc product (GDP) growth and the policy choices that Brazil makes over the coming decades (see Chapterථ9). The gradual weakening of the correlaƟon between rates of GDP growth and energy demand growth is a phenomenon that has been observed in many countries and regions; the extent to which it occurs in Brazil will be condiƟoned by the way that economic acƟvity changes over Ɵme, with shiŌs in producƟvity and in the composiƟon of GDP, and by the way that energy is used to fuel economic acƟvity, as more eĸcient technologies are adopted.1

600

6 000

500

5 000

400

4 000

300

3 000

200

2 000

100

1 000

1990

2000

2010

2020

Billion dollars (2012, MER)

Mtoe

Figure 10.1ථ‫ ٲ‬Brazil GDP and primary energy demand by scenario Current Policies Scenario New Policies Scenario 450 Scenario GDP (right axis)

2030 2035

Note: GDP is expressed in year-2012 dollars in market exchange rate (MER) terms. Mtoe = million tonnes of oil equivalent.

© OECD/IEA, 2013

The implicaƟon of these projecƟons is that Brazil’s energy consumpƟon per capita, which is currently around three-quarters of the global average, rises above world average levels in 2035 in each of the three scenarios. Where these scenarios diīer, though, is in the way that government policies aīect the trajectory of energy demand growth. The fairly limited variaƟon between the Current Policies and New Policies scenarios reŇects the strong dynamics underpinning the rise in energy demand in Brazil and our guarded assessment of 1.ഩ The weakening relationship between GDP growth and energy demand growth projected in our scenarios is a point of divergence with Brazilian energy planning scenarios, where long-term growth rates for the economy and the energy sector are more closely correlated. This is less visible over the Brazilian ten-year planning horizon, but is one factor (alongside different GDP assumptions and other parameters) that leads to divergent projections in the longer term. 330

World Energy Outlook 2013 | Brazil Energy Outlook

the likely impact of the policies so far announced to curb this growth. There remains scope for stronger policy acƟon in some important areas, notably energy eĸciency, which could result in energy demand growing more slowly than in the New Policies Scenario. This is reŇected in an Eĸcient Brazil Case (see Chapterථ12) and also in the 450ථScenario, in which consumpƟon increases by less than 2% per year on average, 0.6ථpercentage points lower than the rate seen in the New Policies Scenarioථ(Tableථ10.1). Table 10.1ථ‫ ٲ‬Brazil total primary energy demand by fuel and scenario (Mtoe) New Policies Scenario

Current Policies Scenario

2011

2020

2035

2020

2035

2020

2035

59

109

141

165

143

174

129

112

3

23

38

77

42

88

32

51

10

15

19

24

19

28

17

17

Natural gas Coal Nuclear

1

4

6

8

6

8

6

11

66

116

148

207

146

204

150

225

Hydropower

18

37

44

58

44

60

44

58

Bioenergy*

48

78

99

138

97

134

102

156

0

1

5

11

5

10

5

11

Renewables

Other renewables Total

138

267

352

480

356

502

334

416

Fossil fuel share

52%

55%

56%

55%

57%

58%

53%

43%

* Includes tradiƟonal and modern biomass uses.

Figure 10.2ථ‫ ٲ‬Primary energy mix in Brazil and the world in the New Policies Scenario 100%

3 4 5 6 7 8 9 10 11

Other renewables Bioenergy

80%

2

450 Scenario

1990 Oil

1

12

Hydro Nuclear

60%

Coal Gas

40%

13

Oil 20%

14 Brazil

World

© OECD/IEA, 2013

2011

Brazil

World

15

2035

The Brazilian energy mix retains its disƟncƟve character in the New Policies Scenario, with the overall shares of fossil fuels and of renewable sources of energy remaining largely unchanged in 2035, compared with 2011 (Figureථ10.2).2 Among the fossil fuels, the share 2.ഩ The remainder of this chapter focuses on the projections for the New Policies Scenario.

Chapter 10 | Prospects for Brazil’s domestic energy consumption

331

16

of oil in Brazil’s primary demand declines (from 41% to 34%), while that of natural gas increases (from 9% to 16%). Coal conƟnues to play only a very small role in Brazil’s energy sector, the share of around 5% being a fracƟon of the global average. Among the renewable sources of energy, the share of bioenergy remains at just under 30% and hydropower falls slightly (from 14% to 12%) while the share of wind and solar, taken together, increases from a very low base to reach 2%. The 43% share in primary energy demand held by renewables in 2035 means that Brazil remains a global leader in low-carbon energy development, well ahead of the world average.

Outlook for the power sector Electricity demand In the New Policies Scenario, electricity demand conƟnues the steep upward trajectory seen in recent years, rising by nearly 3% per year on average over the period from 2011 to 2035, doubling from 471ථterawaƩ-hours (TWh) to 940ථTWh.3 This strong increase in demand Ňows from the growth of Brazil’s economy and the rise in average income levels, driving up electricity consumpƟon in appliances and for cooling. By sector, the largest increase in demand in absolute terms comes from buildings, where residenƟal and services consumpƟon both more than doubleථ(Figureථ10.3); buildings represent more than half of electricity consumpƟon by 2035. Electricity use also grows strongly in industry, by 180ථTWh (see the sector-by-sector discussion for more details). Transmission and distribuƟon (TΘD) losses are relaƟvely high in Brazil, including not only the electricity that is dissipated during t