Climate Policy Integration Into Eu Energy Policy

! Faculteit Economische, Sociale en Politieke Wetenschappen en Solvay Business School – Vakgroep Politieke Wetenschappen

CLIMATE POLICY INTEGRATION INTO EU ENERGY POLICY

CLAIRE DUPONT Promoter: Prof. Dr. Sebastian Oberthür

Proefschrift ingediend met het oog op het behalen van de graad van doctor in de politieke wetenschappen Academiejaar 2012-2013

Cover picture: © Phil Noble/Reuters/Corbis

Printing: Uitgeverij University Press Leegstraat 15 B-9060 Zelzate Tel +32 9 342 72 25 E-mail: [email protected] www.universitypress.be

All rights reserved. No parts of this book may be reproduced or transmitted in any form or by any means, electronic, mechanical, photocopying, recording, or otherwise, without the prior written permission of the author.

For Nico.

Contents

"##)*+)) !''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''!

!1!"#%!"*!"44)"" !''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''!!

!1! "4)!''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''!"!

!1!1+!)!''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''!$!

2&! *!# !''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''!2!

2'2!4"#+!*!''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''!3! 2'3! ;)!))"#;!)# !'''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''!5 ! ! ! &%! !((,%!1% `RaR` # (! T!RRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRR!c! ! 1( !5,!57!(5 5,! `RaRa 1 1!RRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRR!e! 2'4!!))"#;!*)+$5,! !'''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''!21 `RaRb ( 1!RRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRR!g! ! ! !5 5 "(!15(!((1 `RbR` !RRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRR!`Y! 2'5!! !#5(! !)!"*!"!1 !7%!( &7!57!*+!1! (1&!( ;)!*) " !'''''''''''''''''''''''''''''''''''''''''''''''''''''''''!26 `RbRa 1 !RRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRR!`c!

3&!# )$ !"*!4"#+!*!''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''!34!

3'2!##) !"*!*)1 !''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''!35!

! ! &5 !!1% aR`R` # T!RRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRR!ad! ! 1%5,!! &(! aR`Ra (!!RRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRR!ae! aR`RaR`! 30!(EB@A0&9!1@884CC4@9!RRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRR!ag! aR`RaRa! 30!1@E9(47!@1!D30!(EB@A0&9!!94@9!RRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRR!ah! aR`RaRb! 30!(EB@A0&9!&B74&809D!RRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRR!ba! aR`RaRc! 30!(EB@A0&9!1@E9(47!RRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRR!bb! aR`RaRd!D30B!&(D@BC!RRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRR!bc! 3'3!)!! &(!!#" !1)!"*!))+%!#%!*)"))7(1!1(7!( ! ( U% (!''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''!48 aR`Rb (!U S Q S S !RRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRR!bd! ! ! &(!(5%!75% ! !57!U(V( aRaR` Q `hgY !RRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRR!cY! ! 5U%! ( ! !15 (!57!((,%!1%!! &(! aRaRa U Q (! `hhY !RRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRR!cb! 3'4!#!#!&(! !",!!'''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''' !1(7U(!1!1(!57!(5!( !6 6 aRaRb aYYY Q !RRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRR!cf! !

4&!##) !"!1")#!'''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''!68

4'2!##) !"+!#" )!#%! )+" !''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''!68!

! ! !1%!1&((1(!57!175 ! !1%! (,5 bR`R` V !RRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRR!df! ! !("( 5! ! !15 (!1%! (,5 ! bR`Ra V W(X W1X!RRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRR!e`! ! (5 5,! bR`Rb 1!RRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRR!e d ! bR`RbR`!0&CEB492!1!49!D30!A@74(I!AB@(0CC!RRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRR!ed! 4'3! ;)) #"!1")#!'''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''!87 bR`RbRa!0&CEB492!1!49!D30!A@74(I!@EDAED!RRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRR!fc! ! ! &(( 15!57!($55 %!V5(#!! ! (5 !( bRaR` ( !RRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRR!fe ! ! &((!V! !(5! (,5 bRaRa ( !RRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRR!fg ! bRaRaR`!0@1E9(D4@9&74C8!RRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRR!gY! bRaRaRa!4'0B&7!49D0B2@F0B9809D&74C8!RRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRR!ga! bRaRaRb!0G!49CD4DED4@9&74C8!RRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRR!gd! bRaRaRc!5AA74(&D4@9!D@!B0C0&B(3!AB@50(D!RRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRR!gf! ! $55 %!"55U( bRaRb ( !RRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRR!hY ! bRaRbR`!VE9(D4@9&7!49D0BB07&D4@9C!RRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRR!hY! bRaRbRa!@74D4(&7!(@884D809D!RRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRR!ha! bRaRbRb!9CD4DED4@9&7!&9)!A@74(I!(@9D0HD!RRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRR!hb! bRaRbRc!B@(0CC!)4809C4@9!RRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRR!hd! 4'4!##!!''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''!@ 8 bRaRbRd!&40B&B(3I!&8@92!D30!0HA7&9&D@BI!F&B4&'70C!RRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRR!hd!

5&!#! !)!!))#"4)!))+%!#%!'''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''!@@!

5'2!))#"4)!))+%!! ;)!)!!''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''!@@!

! ! 5%! !1%!7("(( cR`R` ( ( !RRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRR!`YY ! 5'3!! )"!+!("((# !!!!)!!,((5!)!#% !!!''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''!217 &(! cR`Ra 7 ( (!!RRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRR!`Ya! ! ! (1&5!V!(5!,! !! &(!1%!1( cRaR` U 1 !RRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRR!`Ye! 5'4!)!! !(1&5!V!(5!,!)!#%!'''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''!21!! &(!1%!! ! @ cRaRa U 1 !RRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRR!`Yf! ! cRbR`!7(1 "(!aYY`VffV(1!! &(! !V!((#5U(!((,%!!1(!V! ((1 1 %! ! (1 "( W(S( 7 X!RRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRR!```! cRbR`R`!1!49!D30!A@74(I!AB@(0CC!@1!D30!aYY`!(S(!74B0(D4F0!RRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRR!``d! cRbR`Ra!1!49!D30!A@74(I!@EDAED!@1!D30!aYY`!(S(!74B0(D4F0!RRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRR!`ab! cRbRa!7(1 "(!aYYhVagV(1!! &(! !V! &(!!(!V!((,%!V!((#5U(! !1(! ! (1 "( W( 7 X!RRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRR!`ae!

cRbRaR`!1!49!D30!A@74(I!AB@(0CC!@1!D30!aYYh!(!74B0(D4F0!RRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRR!`b`! 5'5!)$"+!#!!)!!)!#%!'''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''!251 cRbRaRa!1!49!D30!A@74(I!@EDAED!@1!D30!aYYh!(!74B0(D4F0!RRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRR!`bf! ! ! !1 5! ((5 cRcR` V !RRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRR!`c` ! ! 15!1 ( cRcRa !RRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRR!`ca ! ! ! 5!57!1%!1 ($ cRcRb !RRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRR!`cf! ! 1(!7( cRcRc !RRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRR!` d ` ! 5'6!#!#!"("(#!V!($55!'''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''!26 %!V5(# 7 cRcRd !RRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRR!`db!

6&!#! !)!!))+%!)1"#)!1!4!*+!#%!'''''''''''''''''''''''''''''''!26@!

6'2!))+%!)11#)#%!! ;)!)!!'''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''!271!

! ! (,%!(V51(!V!U!7,!! &(! dR`R` ( (!!RRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRR!`ed! 6'3!! )"!+!(,%!(V51(!V!U!7,!1%5,#!! ;)!)!(!)4!#%!''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''!281 dR`Ra ( !RRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRR!`ef! ! ! (1&5!V!(5!,! !! &(!1%!1( dRaR` U 1 !RRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRR!`fY! 6'4!)!! !(1&5!V!(5!,!)4!#%!''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''!284!! &(!1%!! ! dRaRa U 1 !RRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRR!`f`! ! ! (,%!(V51(!V!U!7,! (1 "(! dRbR` ( 7 WaYYaVh`V(1X!RRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRR!`fd! dRbR`R`!1!49!D30!A@74(I!AB@(0CC!@1!D30!aYYa!(U7!RRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRR!`fg! dRbR`Ra!1!49!D30!A@74(I!@EDAED!@1!D30!aYYa!(U7!RRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRR!`gc! ! &(! !(15 !V! &(! ! dRbRa aY`Y (U7 WaY`YVb`V(1X!RRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRR!`gg! dRbRaR`!1!49!D30!A@74(I!AB@(0CC!RRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRR!`h`! 6'5!)$"+!#!!)!!)4!#%!'''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''!311 dRbRaRa!1!49!D30!A@74(I!@EDAED!RRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRR!`hg! ! ! !1 5! ((5 dRcR` V !RRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRR!aY` ! ! 15!1 ( dRcRa !RRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRR!aYa ! ! ! 5!57!1%!1 ($ dRcRb !RRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRR!aYf! ! 1(!7( dRcRc !RRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRR!a ` a ! 6'6!#!#!"("(#!V!($55!'''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''!32 %!V5(# 7 dRcRd !RRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRR!a`c!

7&!#! !)!!)$ )"!+"!1" !# !)!#%!'''''''''''''''''''''''''''''''''''''''''!32@!

7'2!" !"!+"!! ;)!)!!'''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''!331!

! ! 5 !5!,5!57! !((,%!1% eR`R` (! !RRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRR!aa` ! ! !5 !5!,5!7!1 !1! !57! eR`Ra (! P !RRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRR!aac !

7'3!! )"!+!!5 !5!,5!#!!)!!)$ !V5)"!+"!1" !1 !( !# !)!#%!'''''''''''''''''''''''''''''''''''''''!33@ eR`Rb (! !RRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRR!aaf! ! ! (1&5!V!(5!,! !! &(!1%!1( eRaR` U 1 !RRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRR!aah! 7'4!)!! !#%!!)$(1&5!V!(5!,! )"!+"!1"!! !#&(!1%!! !)!'''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''!347 ! eRaRa U 1 !RRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRR!abY! ! ! (1! %!V!!%!1 ($ !! &(! eRbR` (!!RRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRR!abf! ! 5 !(5! ( #!V! (,%! eRbRa S( ( W (S(X!RRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRR!abh! eRbRaR`!1!49!D30!A@74(I!AB@(0CC!@1!D30! (S(!2E4)07490C!RRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRR!acd! eRbRaRa!1!49!D30!A@74(I!@EDAED!@1!D30! (S(!2E4)07490C!RRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRR!ach! ! !(5! (,%! ,5(!V! (1"(%! eRbRb ( ( W((X!RRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRR!adY! eRbRbR`!1!49!D30!A@74(I!AB@(0CC!@1!D30!aYYh!((!02E7&D4@9!RRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRR!adb! 7'5!)$"+!#!!)!!)$ )"!+"!1" !# !)!#%!'''''''''''''''''''''''''''''''''''''''!367 eRbRbRa!1!49!D30!A@74(I!@EDAED!@1!D30!aYYh!((!02E7&D4@9!RRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRR!add! ! ! !1 5! ((5 eRcR` V !RRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRR!adf ! ! 15!1 ( eRcRa !RRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRR!adh ! ! ! 5!57!1%!1 ($ eRcRb !RRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRR!ae`! ! 1(!7( eRcRc !RRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRR!a e e ! 7'6!#!#!"("(#!V!($55!'''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''!37 %!V5(# @ eRcRd !RRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRR!aef!

8&!##!!'''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''!382!

8'2!))"#;!"!"*!*)+!'''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''!382 ! 8'3!"!))"#;!1*+!''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''!386 ! ! ! (! !V! &(!55%!V! &(!("(!V! fRaR` 1!RRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRR!afe! ! 7,!! &(!515 !V! &(!($55 %!V5(# fRaRa V !RRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRR!afh ! fRaR`R`! 30!B0(@294D4@9!W&9)!&BD4(E7&D4@9X!@1!1E9(D4@9&7!49D0BB07&D4@9C!&C!&!14BCDS@B)0B! F&B4&'70!1@B!1!RRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRR!agY! fRaR`Ra!@74D4(&7!(@884D809D!&C!&9!49CE114(409D!0HA7&9&D4@9!1@B!70F07C!@1!1!RRRRRRRRRRRRRRRRRRRR!agb! fRaR`Rb!9CD4DED4@9&7!&9)!A@74(I!(@9D0HD!F&B4&'70!AB@F4)0C!D30!'&(62B@E9)!(@9D0HD!RRRRRRR!agd! fRaR`Rc!UB0&6492!)@G9!D30!AB@(0CC!)4809C4@9!D@!0HA7&49!1!49!D30!A@74(I!@EDAED!RRRRRRRRR!aha! 8'4!!"%!1!1*+!''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''!3@6 fRaR`Rd!&40B&B(3I!@1!0HA7&9&D@BI!F&B4&'70C!RRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRR!ahd! 8'5!# 4! ! ! )" !)!'''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''!3@7 ! 8'6!1! !)!))"#;!'''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''!3@9 !

)+!)&!)"!"*!)#)*" !1!#%")!'''''''''''''''''!412!

44+";%!'''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''!414

")$!&! !1!)+" !''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''!444!

")$!&!)+"! )$ !'''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''!448!

)" %!) "4;+! ;)!)!)"!#! %!) )#0&!'''''''''''''''''''''''''''''''''''''''''''''''''''''!448! )" %!! ;)!1!# +!1! ;)!)!)"!!!) 1)!0&!'''''''''''''''''''''''''''''''''''''''''''!451 ! !)!1!#)*!)!1! ;)!)!)"!"") !'''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''!456 !

")$!&! !1! )")#!''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''!458!

Acknowledgements

There are a great many people who have encouraged, helped and motivated me along the way to the completion of this PhD thesis, and I would like to extend my gratitude to all of them.

The project would not have been possible without the support and guidance of my supervisor, Sebastian Oberthür. His enthusiasm for my work never seemed to fade, and his well-timed encouragements, criticisms and comments served to keep me on the straight path! I am very grateful for all the time and effort he has spent working with me over the years. Harri Kalimo and Ingmar von Homeyer deserve my sincere gratitude as active members of my doctoral committee. Both Harri and Ingmar could be relied on to raise challenging questions on my research, which inevitably led to interesting conversations and considerable improvements in my work.

The Institute for European Studies has been my second home since 2008. The working environment and the facilities the Institute provides are top rate. But it is the people at the IES that help create the wonderful working environment, with plenty of humour, mutual support and enjoyable coffee breaks. A special mention goes to the (past and present) members of the Environment and Sustainable Development cluster at the IES – Armelle, Alex, Eléonore, Ernesto, Florian, Harri, Ioannis, Johanna, Justyna, Katja, Laura, Lisanne, Max, Olof, Radostina, Sebastian, Thomas, and Thuy – but I want to send thanks to all my colleagues at the IES for making the Institute a friendly and fun place to work. I would especially like to thank Hilde, who always has the answers to my questions. I would also like to thank Chad Damro, at the University of Edinburgh, for welcoming me as a visiting researcher at the Europa Institute during the summer of 2010. I very much enjoyed my stay at the Institute and was delighted to exchange ideas and receive feedback from Chad.

This project would not have been completed without the support of friends and family in Belgium and abroad. I am eternally grateful to my excellent team of proofreaders for helping me in the final moments of my dissertation preparation: Aedín, Anna, Anne, Caroline, Daniel, Niamh, Nico and Terry. I wish to thank my parents, Marie and Hubert – who often wondered what exactly I was doing all this time, but never failed to support me. Also, I thank my siblings and their better halves, Ed and Tina,

Lorcan and Denise, Niamh and Nathan, and Hugh and Alicia, for supporting me in my work, but also for playing the devil’s advocate so often that I am now well- equipped to respond to any climate sceptics or critics! Many thanks to Sylvie, André and Christophe for their interest in my work and for giving me plenty of support in the last few years of the project. Special thanks must also go to friends around the world: Terry, Niamh, Aedín, Caroline, Aine, Lucy, Paudie, Ross, Neil, Jillian, John, Bahija, Sogol, Bart, Emilie, Sylvie, Yohann, Naghmeh, Rachel, Saba, Hushi, Ben, Keri, Marie, Steve, Rusty, Helen, Fiona and many others who distracted me and supported me from near and afar.

I could not have made it without Nico. He has constantly supported and motivated me, encouraged me when I was down, and helped me relax and enjoy life away from work. To make sure I had the best environment in which to complete my PhD, he moved to Brussels and left behind the life he had built in one of his favourite cities. “Thank you” simply cannot cover it! I dedicate this dissertation to him.

List of acronyms and abbreviations

ALDE Alliance of Liberals and Democrats for Europe

BAU Business-as-usual bcm Billion cubic metres

BP British Petroleum

BPIE Buildings Performance Institute Europe

°C Degrees Celsius

CAIT Climate analysis indicator tool

CCPM Common and coordinated policy and measure

CCS Carbon capture and storage

CDM Clean development mechanism

CFC Chlorofluorocarbon

CH4 Methane

CO2 Carbon dioxide

COP Conference of the Parties

COR Committee of the Regions

COREPER Committee of Permanent Representatives

CPI Climate policy integration

DG Directorate General

EAP Environmental action programme

EC European Community

ECCP European climate change programme

ECF European Climate Foundation

ECSC European Coal and Steel Community

EEA European Environment Agency

EEB European Environmental Bureau

iii

EEC European Economic Community

EEPR European energy programme for recovery

EESC European Economic and Social Committee

ENVI Environment committee in the European Parliament

EPB Energy performance of buildings

EPBD Energy performance of buildings directive

EPI Environmental policy integration

EPP European People’s Party

EREC European Renewable Energy Council

ETS Emissions trading scheme

EU European Union

Euratom European Atomic Energy Community

EWEA European Wind Energy Association

GHG Greenhouse gas

HFC Hydrofluorocarbon

IEA International Energy Agency

IPCC Intergovernmental Panel on Climate Change

ITGI Italy-Greece-Turkey interconnector

ITO Independent transmission operator

ITRE Energy committee in the European Parliament

KP Kyoto Protocol kWh Kilowatt hour

LNG Liquefied natural gas

MEP Member of the European Parliament

Mtoe Mega tonnes of oil equivalent

MW Megawatt

NF3 Nitrogen trifluoride

N2O Nitrous oxide

NGO Non-governmental organisation

OECD Organisation for Economic Cooperation and Development

PFC Perfluorocarbon

QMV Qualified majority voting

RE Renewable energy

RES-E Renewable energy sources of electricity

S&D Group of the Progressive Alliance of Socialists and Democrats

SEA Single European Act

SF6 Sulphur hexafluoride

TAP Trans-Adriatic pipeline tcm Trillion cubic metres

TEC Treaty establishing the European Community

TEN-E Trans-European networks for energy

TFEU Treaty on the functioning of the European Union

UK United Kingdom

UNEP United Nations Environment Programme

UNFCCC United Nations Framework Convention on Climate Change

US United States of America

WMO World Meteorological Organisation

WTO World Trade Organisation

WWF World Wide Fund for Nature

List of Tables

Table 1.1 Scale to measure the level of CPI in the policy output. 7

Summary of case studies, showing variation on the dependent Table 1.2 13 variable as a main case selection criterion.

Some of the main policy developments in EU energy and Table 2.1 41 climate policy, 1950s-1980s.

Some of the main policy developments in EU energy and Table 2.2 climate policy in the 1990s (with agreed Directives, Decisions 44 or Regulations in italics).

Some of the main policy developments in EU energy and Table 2.3 climate policy in the 2000s (with agreed Directives, Decisions 48 or Regulations in italics).

Table 3.1 Indicators to measure the level of CPI in the policy process. 73

Table 3.2 Scale establishing the extent of CPI in the policy output. 75

Summary of some of the main approaches and variables in Table 3.3 77 EPI literature.

The nature of functional interrelations and their potential Table 3.4 92 effect on CPI.

Scale measuring political commitment to, first, combating Table 3.5 93 climate change, and, second, advancing CPI.

Summary of explanatory framework, as derived from theories Table 3.6 98 of European integration and EPI literature.

Percentage share of RE in final energy consumption in EU27 Table 4.1 105 and in each member state in 2004 and 2010, with 2020 target.

Levels of CPI in the policy process and output of the 2001 Table 4.2 140 RES-E and the 2009 RE Directives.

Summary of the explanatory variables for CPI in the 2001 Table 4.3 154 RES-E Directive and the 2009 RE Directive.

Table 5.1 Main pieces of EU legislation promoting energy efficiency. 164

Levels of CPI in the policy process and output of the 2002 Table 5.2 201 and 2010 EPBDs.

vii

Summary of explanatory variables for CPI in the 2002 EPBD Table 5.3 215 and the 2010 EPBD recast.

EU gas import capacity 2010-2020, taking account of projects Table 6.1 under construction, but excluding proposed projects such as 229 the Southern gas corridor.

Comparison of gas import infrastructure capacity to natural Table 6.2 236 gas consumption expectations under very high CPI scenarios.

Levels of CPI in the policy process and output of the external Table 6.3 gas infrastructure policy of the 2003 and 2006 TEN-E 257 Decisions and the 2009 EEPR Regulation.

Summary of explanatory variables for CPI in external gas Table 6.4 infrastructure policy in the 2003 and 2006 TEN-E Decisions 268 and the 2009 EEPR Regulation

Summary of levels of CPI in the policy process and output of Table 7.1 277 the case studies, compared to initial expected results.

Summary of explanatory variables for levels of CPI in the Table 7.2 280 case studies.

Extent of CPI in the policy process and output of the cases, Table 7.3 the nature, and the recognition of the functional interrelations 282 in the policy process (part of the process dimension variable).

Extent of CPI in the policy process and output, and levels of Table 7.4 political commitment to combating climate change, generally, 284 and to advancing CPI, in each case.

Extent of CPI in the policy process and output, and the role of Table 7.5 the institutional and policy context in explaining the levels of 291 CPI.

CPI in the policy output, compared to the three indicators and Table 7.6 293 the overall level of CPI in the policy process.

viii

List of Figures

Figure 2.1 The policy cycle. 26

Figure 2.2 GHG emissions in the EU15 and EU27, 1990-2010. 39

Figure 3.1 Increasing scale of policy coordination. 59

Percentage share of renewable energy in final energy Figure 4.1 103 consumption in the EU27, 2004-2010.

Very high or complete CPI from 2000-2010, compared to Figure 4.2 BAU scenarios from 2006 and from 2011 (measured as a 109 percentage share of final energy consumption).

Policy output of the 2001 RES-E Directive compared to Figure 4.3 BAU and to very high levels of CPI (measured as a 124 percentage share of gross inland energy consumption).

CPI in the policy output of the 2001 RES-E Directive, Figure 4.4 assuming lower advances in RE share between 2000 and 126 2010 than between 2010 and 2050.

CPI in the policy output of the 2009 RE Directive, Figure 4.5 compared to BAU scenarios and the very high CPI 138 trajectory, from 2005 levels.

CPI trajectory for the RE share to 2050, with lower Figure 4.6 139 expectations to 2020.

Development of the energy intensity of the economy of the Figure 5.1 EU27 between 2001 and 2010, measured in kilograms of oil 160 equivalent per EUR 1000, with (dotted) trend line.

Final energy consumption in the EU27 from 2000 to 2010, Figure 5.2 measured in Mega tonnes of oil equivalent (Mtoe), with 161 (dotted) trend line.

Evolution in housing floor area and energy consumption for Figure 5.3 166 space heating in the EU27, 1990-2008.

Very high/complete CPI trajectory for energy consumption in buildings in the EU27 from 2000 to 2050, compared to Figure 5.4 the BAU trajectory (that includes actual levels of 173 consumption in buildings from 2000 to 2010; measured in Mtoe).

CPI in the policy output of the 2002 EPBD, with measures expected to avoid the consumption of 55 Mtoe of energy in Figure 5.5 185 buildings in the EU between 2000 and 2010 (compared to BAU and to the very high CPI range outlined in Figure 5.4).

EPBD measures compared to actual consumption of energy Figure 5.6 186 in buildings, 2000-2010 (measured in Mtoe).

CPI in the policy output of the 2010 EPBD recast, with measures expected to avoid between 60 and 80 Mtoe of Figure 5.7 199 energy consumption in buildings in 2020 (compared to BAU and very high/complete CPI range; measured in Mtoe).

Updated BAU scenario from 2011, which includes expected EPBD impact on the energy consumption of buildings to Figure 5.8 200 2020, compared to the 2006 BAU trajectory and CPI ranges to 2020 (measured in Mtoe).

Evolution of natural gas consumption and production in the Figure 6.1 EU from 2000 to 2010 (measured in billion cubic metres, 225 bcm).

Share of LNG supplies from different source countries into Figure 6.2 226 the EU in 2010 (measured in bcm).

Range of very high or complete CPI gas consumption trajectories in the EU27 to 2050, compared to the BAU and Figure 6.3 234 decarbonisation scenarios for the Commission’s Energy Roadmap (measured in bcm).

Very high or complete range of CPI to 2050, with continued natural gas consumption to 2015 before consumption begins Figure 6.4 235 to reduce by 2020 in line with decarbonisation scenarios (measured in bcm).

Introduction

Climate change is often considered a “wicked” problem (Jordan, Huitema, & van Asselt, 2010, pp. 4-5). It is a cross-cutting, long-term, global issue that presents policymakers with many challenges in their efforts to respond to the issue. The European Union (EU) has developed a leadership stance on the climate issue over time, and aims to advance global agreement through diplomacy and credible leadership by example (Wurzel & Connelly, 2011b). Ensuring that its domestic policies are sufficiently ambitious to demonstrate to the global community the potential and importance of advancing policies to combat climate change is one of the EU’s leadership strategies (Oberthür & Roche Kelly, 2008).

The focus of this research project is on the EU’s internal energy policies and on whether climate policy objectives are sufficiently integrated into these policies. This dissertation reports on a study of climate policy integration (CPI) into three cases of EU energy policy. I describe and analyse the level of CPI in EU energy policy between 2000 and 2010, with the aim of explaining why varying levels of CPI within the EU’s energy policy exist (as a single policy sector). This research falls firmly within, and adds to, academic literature on environmental and climate policy integration (EPI/CPI; Adelle & Russel, 2013; Jordan & Lenschow, 2010; Lafferty & Hovden, 2003; Lenschow, 2002b; Nilsson & Eckerberg, 2007). The study is an in- depth case study analysis of EU energy policies, and a testing ground for operationalising CPI.

This dissertation contributes to the advancement of the academic literature on EPI and CPI by presenting a conceptual framework that may be used to assess and explain the levels of CPI in both the process and output of EU policies. Understanding is advanced through in-depth empirical research on the integration of the complex problem of climate change into EU policy. In this study, I highlight a combination of explanatory variables that are most valuable for understanding the extent of CPI in policymaking. The results provide valuable reflection, not only for academic research

1 Introduction on EPI and CPI, but also for policymakers engaged in the climate and energy policy sectors.

In this introduction, I first describe the background behind the research project, referring to the complexity of responding to climate change, and policy developments in the EU. In the second section, I present the research set-up, including the central question of the project. I also describe the research design and methods used. Third, I present the structure and main points of each of the following chapters, including a summary of the main findings and conclusions.

1.1 Background Climate change is one of the greatest threats facing humanity in the twenty-first century. It is more than an environmental issue. With the effects of climate change expected to increase pressure on our food and energy systems, combating climate change (and adapting to its impacts) is a matter of survival. The fourth assessment report of the Intergovernmental Panel on Climate Change (IPCC) highlighted, with high levels of certainty, that climate change is caused by human activity – and most importantly by the burning of fossil fuels (such as coal, oil and gas), which emits dangerous greenhouse gases (GHGs). Over time, these GHGs accumulate in the earth’s atmosphere, which results in less heat escaping into space and an increase in global temperature (the “greenhouse effect”; IPCC, 2007). In this same report, the IPCC described the challenge for policymakers to respond to climate change. It is a global crisis that cuts across frontiers, affects developed and developing countries, and requires action and policy measures across a wide variety of policy fields. Policy sectors that affect climate change, or that will be affected by climate change, include agriculture, biodiversity, energy, fisheries, health, industry, transport, waste management, water, among others. Effectively combating climate change, as highlighted by the IPCC, means ensuring that climate change is “integrated”1 into the policy process and policy output2 of each of these policy sectors.

1 Policy integration is at times considered synonymous to “mainstreaming”, which means including a certain policy’s objectives in the development and output of other policies. For more on the conceptualisation of the term “policy integration”, see chapter three.

2 Throughout this dissertation, I refer to the policy process, namely the negotiations among policymakers leading to agreement on a policy instrument, and the policy output, namely the agreed policy (e.g. a piece of legislation laying down certain measures). The policy output should be distinguished from the policy outcome, which is not under investigation here. The outcome refers to the actual effect of a policy measure’s implementation (see chapters two and three for more detail).

2 Introduction

As a “wicked problem” (Jordan, Huitema, & van Asselt, 2010, pp. 4-5), climate change belongs to a set of problems that “challenge established social values and institutional frameworks, defy analysis, and have no obvious solutions” (ibid.). While such a framing may seem defeatist, it is certainly true that, despite growing scientific evidence, humanity has nonetheless continued its fossil fuel-consuming development path that upsets the precarious balance in the climate system (IEA, 2011a, 2013).

Despite these challenges, the EU has developed several policy measures over the years in its effort to exert international leadership in the fight against climate change. Since the early 1990s, the EU has expressed a desire to lead the international community in solving this issue and it has increasingly added credibility to its rhetoric by agreeing internal policy measures to reduce its own emissions of GHGs (see chapter two; Gupta & Grubb, 2000; Oberthür & Roche Kelly, 2008; Wurzel & Connelly, 2011). EU citizens generally view climate change as an important issue, and actions to combat climate change in the EU have received wide popular support (Eurobarometer, 2008, 2011). These realities mean that climate change has enjoyed a higher status than many other environmental issues in the EU – to the extent that the Commission gained a Directorate General (DG) and a Commissioner devoted solely to the issue of climate change in 2010 (Commissioner Connie Hedegaard).3 Given the complex reality of the climate issue, however, it is clear that environmental policies alone are insufficient to address the problem. With many other sectors causing climate change, responses need to be developed in these sectors in particular. CPI, then, represents one possible policy strategy that can be deployed to respond to the challenge of making policy to combat climate change.

But what is meant by climate policy integration? In some respects, climate policy integration is an example of environmental policy integration (EPI). EPI is a legal objective of the EU. Article 11 of the Treaty on the Functioning of the European Union (TFEU) states: “Environmental protection requirements must be integrated into the definition and implementation of the Union’s policies and activities, in particular with a view to promoting sustainable development”. This Article (formerly Article 6 of the Treaty establishing the European Community, TEC) appeared in its first formulation in the Single European Act (SEA) of 1987, which stated in its Article

3 See: http://ec.europa.eu/dgs/clima/mission/index_en.htm, accessed: 4 April 2013. See also chapter two.

3 Introduction

130r(2): “environmental protection requirements shall be a component of the Community’s other policies”. EPI thus has a relatively long history in the EU, but what precisely EPI is in practice has also long been discussed. Whether EPI is considered a policy or legal principle (Nollkamper, 2002), an objective for improving the efficiency of the policy process or a requirement for policies to improve the state of the environment (a policy outcome; Persson, 2004), policymakers and scholars have not managed to agree. At the very minimum, “integration” implies that one policy’s objectives become part of another policy’s development (see chapter three; Briassoulis, 2005b).

1.2 The research project With the adoption of the 2009 “integrated” package of policy measures on climate and energy, it seemed not only that the EU had finally followed through on its promises of climate leadership, but that it had also practised EPI (Adelle, Pallemaerts, & Chiavari, 2009; Adelle, Russel, & Pallemaerts, 2012; ENDS Europe, 2008g; Oberthür & Roche Kelly, 2008; Wurzel & Connelly, 2011). This development also raised several questions that form the central motivations for this research project:

• Just how “integrated” are climate policies into EU energy policies anyway? • Has the level of CPI in EU energy policy changed over time? • Is CPI uniform across energy policies? • How could potential variances in the levels of CPI be explained?

The research aim of this dissertation can be summed up as a two-step research question:

What is the extent of climate policy integration into the EU’s energy policy, and why?

1.2.1 Why EU energy policy? The energy sector is one of the biggest contributors of GHG emissions in the EU. Whether it is through the production, transmission or consumption of energy, GHG emissions from energy overall are estimated to be at the source of about 80 per cent of total GHG emissions in the EU (EEA, 2010, p. 31; European Commission, 2011b). Mitigating (or combating) climate change requires the reduction of GHG emissions as

4 Introduction soon as possible, and by 2050 to a level 80 to 95 per cent lower than the level of GHG emissions in the EU in 1990. This 80 to 95 per cent reduction in GHG emissions in the EU is a commitment made by the European Council in October 2009 (European Council, 2009b). It represents a translation of scientific estimates of the effort required to avoid dangerous anthropogenic interference with the climate system, and to ensure that global temperature increase does not exceed two degrees Celsius (see European Commission, 2007e; IPCC, 2007). 4 This 2050 target to reduce GHG emissions by 80 to 95 per cent in the EU is thus both a scientific target and a political commitment. As can be seen in Figure 2.1 (chapter two), the EU is already on its way to achieving reductions in its GHG emissions compared to 1990 levels, but the remaining effort required is significant. It would not be possible to meet the 2050 objective without adopting policy measures in the energy sector that ensure the reduction of GHG emissions.

In 2011, the Commission released its “roadmap for moving to a competitive low carbon economy in 2050” (European Commission, 2011b), which recognised that the greatest reductions in GHG emissions will need to occur in sectors that produce and consume energy (ibid., p. 6). Certain sectors of the economy will face greater obstacles in reducing their GHG emissions (agriculture, for example). This communication was followed by a detailed roadmap for the energy sector (called the “energy roadmap 2050”; European Commission, 2011a), which outlines a number of scenarios for the energy sector’s contribution to the EU’s effort to reduce GHG emissions. This effort amounts to at least an 85 per cent reduction in GHG emissions from the energy sector, according to the Commission (European Commission, 2011c, p. 2).5 It is thus clear that aspirations exist within the EU to ensure that climate policy objectives are well integrated into EU energy policy (as also demonstrated by the above-mentioned 2009 climate and energy package of policy measures). The energy sector is a crucial sector for combating climate change – is CPI in the energy sector sufficient to meet the long-term climate policy objectives? The energy sector should

4 While some commentators argue that even a two-degree temperature increase is too high (see, for example, comments from James Hansen on the potential disastrous effects of a two-degree temperature rise, http://www.livescience.com/17340-agu- climate-sensitivity-nasa-hansen.html, accessed: 11 May 2013), I rely in this study on the confirmed, peer-reviewed research reported by the Intergovernmental Panel on Climate Change (IPCC).

5 Other scenarios argue for a complete decarbonisation of the energy sector (e.g. EREC & Greenpeace, 2010).

5 Introduction provide rich empirical data for the further development of the conceptual framework used in this project.

1.2.2 Conceptualising and operationalising CPI For the purposes of this research project, CPI is understood as a matter of degree, and as a policy tool or strategy to ensure that climate protection requirements are integrated into both the policy process and policy output (the final agreed text) of, in this case, EU energy policy. This approach provides a methodological advantage in that I can operationalise CPI to take account of lower and higher levels of CPI. However, such an operationalisation requires a benchmark against which the empirical reality can be measured. Thus, the normative stance of “principled priority” (Lafferty & Hovden, 2003) for environmental/climate policy objectives in non- environmental sectors can represent the highest point of CPI – a benchmark that can be established in reference to long-term climate policy objectives to reduce GHG emissions in the EU by between 80 and 95 per cent by 2050, compared to 1990 levels (European Council, 2009b). With CPI being understood as the integration as climate policy objectives into the process and output of policies, the actual level of CPI can be captured when measured against the benchmark very high standard of CPI. Understanding CPI as a matter of degree (Bryner, 2012) has the advantage of capturing nuances in the empirical realities of CPI – it is not a matter simply of whether CPI exists or does not exist.

Measuring CPI in the policy output means, first, measuring the gap between the business-as-usual (BAU) scenarios6 for that policy area and the very high levels of CPI (the benchmark). Second, the level of CPI for a particular policy output is established by measuring how much of this gap is closed by the output of the policy measure under examination. Accordingly, a fivefold scale can be used to establish the level of CPI, from none or very low over low, medium, high to very high (see Table 1.1 below; also Table 3.2).

6 Business-as-usual (BAU) scenarios are used throughout the empirical chapters four to six for the purposes of measuring the level of CPI in the policy output. A BAU scenario is an estimate of the future scenario in a particular policy sector had no policy measure been adopted.

6 Introduction

No/very Very high/ Low Medium High CPI in the low CPI complete policy output: 0-20% 21-40% 41-60% 61-80% 81-100%

Table 1.1: Scale to measure the level of CPI in the policy output.

A similar fivefold scale for measuring the level of CPI in the policy process is applied in this project. In the case of the policy process, however, there are three main factors that are measured. First, the involvement of policymakers internally to the EU institutions who could be likely to advance climate concerns is measured (called in this research project “internal pro-climate stakeholders”).7 This includes measuring the involvement of, for example, DG Environment (or later, DG Climate Action) in the European Commission (hereafter: Commission); the environment committee of the European Parliament (hereafter: Parliament); and the environment formation of the Council of ministers (hereafter: Council) in the policy process. Second, the involvement of external stakeholders that promote action on climate change is also important for measuring the level of CPI in the policy process. These stakeholders include environmental non-governmental organisations (NGOs) and industries that benefit from strong climate policies (including the renewable energy industry, industries promoting environmentally-friendly products, industries providing services for improving energy efficiency, among others). Finally, it is also crucial that those involved in the policy process recognise the functional interrelations between the policy measure being negotiated and climate policy objectives. In other words, do policymakers explicitly recognise and consider how the policy measure they are negotiating affects or is affected by objectives to reduce GHG emissions to 2050? In Table 3.1, I outline specific indicators for each of these three factors to provide an overall level for CPI in the policy process. This level is again measured on a scale of none/very low over low, medium, high to very high (see chapter three for a more in- depth discussion on the conceptualisation and operationalisation of CPI).

7 A “stakeholder” is a person or organisation with an interest in an issue or topic. The Oxford English Dictionary emphasises the link to business language, in that it defines a stakeholder as someone that is “invested” in a particular (business or) issue.

7 Introduction

1.2.3 Explaining CPI Once the level of CPI is measured, how can it be explained? The literature on EPI presents a myriad of explanatory factors, and which factors are highlighted often depends on the understanding of EPI of the author (whether as a legal or political principle, an aspirational goal, a policy objective, a tool for improving the policy process, a requirement for actual benefits for the environment in implementation, or other). In chapter three, I present a review of these different explanatory factors and compare them with factors derived from theories of European integration (and especially neofunctionalism, liberal intergovernmentalism and new institutionalism) that can explain the EU policymaking process. Along the lines of Wiener and Diez, I therefore understand each of these theories as part of a mosaic, or part of a jigsaw puzzle, that provides for complementary understanding of EU processes towards advancing the European integration project (Diez & Wiener, 2009). It should be made clear, however, that these theories serve the purpose of inspiring the explanatory framework for this research project, and are not tested here for their validity.

The literature on EPI and European theories of integration together suggest four main potential explanatory variables (independent variables) that could explain the (variance in) levels of CPI in the policy processes and outputs of EU energy policies (the dependent variable). These factors are:

• The nature of the functional interrelations between long-term climate policy objectives (i.e. to 2050) and the policy objectives of the policy sector under investigation; • The level of political commitment demonstrated by the EU (including the EU leaders and its institutions) to, first, combating climate change in an overarching sense, and, second, to advancing levels of CPI in sector policymaking; • The institutional and policy context (including decision-making rules and procedures internally to the EU, and external events or shocks, such as the economic crisis or the agenda of the international climate negotiations); • And, the process dimension (namely the level of CPI in the policy process), which can help explain the level of CPI in the policy output only.

8 Introduction

The first three of these explanatory factors may help explain the level of CPI in the policy output and the policy process, while the fourth factor (being the process itself) can only help explain the level of CPI in the policy output. This explanatory framework is applied to the results of the analysis of CPI in the EU’s energy policy. In chapter seven, I discuss the weight of the variables and pinpoint some of the crucial variables for the advancement of CPI.

From the EPI literature, one hypothesis that can be derived is that political commitment affects the advancement of CPI, and can be considered to play a critical role (Jordan & Lenschow, 2010). I would add the hypothesis that political commitment, both to combating climate change and to advancing CPI in sectoral policymaking is an insufficient enabler of CPI. Thus, I expect to find evidence that a combination of variables is required for advancing CPI. Advancing CPI throughout the policy process and in the policy output in the EU involves many actors, with policy not being made in a vacuum, but in the context of external events and institutional procedures. As such, political commitment may be important for an initial push to advance CPI in the process, but such commitment is likely to require sustained effort in the face of many actors and contexts throughout policy negotiations. If this hypothesis is confirmed through the empirical analysis, the combination of variables that do ensure CPI may potentially be identified.

Additionally, I would expect to find that the nature of the functional interrelations between climate policy objectives and the policy sector, and especially the recognition (as part of the process) of these interrelations over a long-term horizon to 2050, would play a role as a first-order explanatory variable. The nature of these interrelations (if they are clear, close, direct and obvious and/or if they are synergistic) and the recognition of these interrelations may already advance CPI from the first moments of the policy process (see chapter three for a more detailed discussion of the explanatory framework). These hypotheses played an initial role in the case selection process, as described in section 1.3.1 below, in that they helped highlight the variation that could be expected on the dependent variable for each case in a first survey of the potential cases of interest (see below).

9 Introduction

1.3 Research design This research project is of primarily a qualitative nature, although certain parts of the research contain some quantitative elements (for example, in terms of analysing statistics on GHG emissions, energy consumption and production or renewable energy generation). The main research strategy deployed involves analysing a limited number of case studies of EU energy policy to establish the levels of CPI in their policy processes and outputs (George & Bennett, 2004; Gerring, 2007; King, Keohane, & Verba, 1994; Swanborn, 2010; Yin, 2003, 2009). Analysis in the case studies follows a process tracing strategy (Bennett, 2008; Checkel, 2008), including several data-collection and analysis techniques: document analysis, literature review, and complementary semi-structured interviews (Hopf, 2004; Kvale & Brinkman, 2009; Rathbun, 2008).

1.3.1 Qualitative case selection To answer the research question (what is the extent of climate policy integration into EU energy policy, and why?), I selected a number of cases of EU energy policy to examine in detail. The universe of cases from which I chose included all EU energy policies. After a survey of this universe of cases highlights that EU energy policies, I found that they could be categorised into three main types (linked to the categories of EU competence on energy policy in Article 194(1), TFEU):

(1) EU internal policies linked to the production and transmission of energy; (2) EU internal policies linked to the consumption of energy; (3) EU external policies linked to ensuring continued supplies of energy to the EU.

The first category of energy policies includes policy measures on, for example, renewable energy generation, the internal market for energy, and on coal and carbon capture and storage (CCS) technologies. The second category of energy policies includes, for example, policy measures to reduce energy consumption by improving the energy-use of products, energy labelling and the energy performance of buildings, among others. Finally, the third category of policies is related to external relations. This category deals specifically with ensuring the security of energy supplies, for which only limited competence exists at the EU level. Bilateral negotiations for supplies of energy are often in the hands of member states, but the EU itself is also involved in energy dialogues with key energy partners, such as Russia (Hadfield,

10 Introduction

2008; Piket, 19-20 June 2003; interview 9). Other policy measures at EU level in this third category include, for example, policies on stocks of natural gas and oil, policies in response to energy supply emergencies and policies on infrastructure measures for imports of energy sources (under the trans-European network for energy guidelines, for example, or the European energy programme for recovery).

For this research project, I followed what is known as a most-similar case selection strategy, where cases are chosen on the basis of variation on the variables of interest (Gerring, 2007; 2008, p. 668; Yin, 2009). Thus, the case studies chosen vary in terms of the dependent variable (the level of CPI). Some variation on the independent variables (namely, functional interrelations; political commitment; institutional and policy context; and the process dimension) is thus also expected, where the variation of the independent variables explains the variation on the dependent variable (King, et al., 1994, pp. 91-93).

Nevertheless, the case studies are bounded within a specific universe (Klotz, 2008; Rohlfing, 2012, pp. 24-28). First, the cases chosen are geographically bounded, in that I focus on EU energy policies that have effect internally to the EU. Second, the cases are institutionally bounded, in that the cases chosen were all agreed under the co-decision procedure8 of the EU (with the Commission proposing legislation and the Parliament and Council co-deciding, see also chapter two). Third, the case studies are substantively bounded as they are all drawn from the energy policy domain. Fourth, the cases are temporally bounded: for this research project, the timeframe of analysis is 2000-2010. In order to see any developments over time, I chose cases with policies that had developments in the early and late years of this decade. Additionally, and within these boundaries, I aimed for a selection of cases that drew from as many types of EU energy policy as possible generally, while still ensuring the number of cases was limited enough to ensure manageable and in-depth research. I therefore chose one case from each of the three categories mentioned above.

The first case study chosen, described in chapter four, is on EU renewable energy (RE) policy. Two policy instruments developed between 2000 and 2010 form the centre of analysis for this case study: the 2001 Directive 2001/77/EC on the

8 Renamed the “ordinary legislative procedure” in the TFEU.

11 Introduction promotion of electricity produced from renewable energy sources (RES-E Directive) and the 2009 Directive 2009/28/EC on the promotion of the use of energy from renewable sources (RE Directive). The RE case is expected to demonstrate high levels of CPI in the policy process and output in the later half of the 2000s, especially given the synergistic interrelations between climate policy objectives to reduce emissions of GHGs (especially from fossil fuels) and the objectives of RE policy (to increase the share of RE in the EU). This expected result for RE policy is expected to develop over time from more medium levels in the early 2000s. This assumption is based on the fact that the 2001 RES-E Directive resulted in indicative targets for the share of RE in the EU, while the 2009 Directive included binding targets. The hypothesis that the nature of the interrelations between the policy sector and climate policy objectives is a crucial first-step factor for any demand for CPI to materialise also inspired this estimate of results for RE policy (see section 1.2.3 above).

The second case study, discussed in chapter five, comes from the second category of EU energy policy measures and is on the energy performance of buildings (EPB). Again, there are two policy instruments under focus in this case: the 2002 Directive 2002/91/EC on the energy performance of buildings and its 2010 recast Directive 2010/31/EU. This case is expected to produce different results on the dependent variable – with more medium levels of CPI expected overall in the policy process and output of the 2010 Directive, evolving from low to medium levels in the 2002 Directive. Given the synergistic interrelations between reducing energy consumption and reducing GHG emissions, the medium levels expected are more closely related to issues of political commitment to agree on ambitious action to improve the energy performance of buildings. For both the 2002 and 2010 Directives, non-binding energy efficiency targets were in place, which hints at lower commitment to achieving the goals of these policies. This sector has historically faced problems of implementation and ambition (see chapter five; Boasson & Wettestad, 2013).

Finally, the third case study, linked to the third category of EU energy policy measures listed above, relates to policies that promote the development of importing natural gas infrastructure (see chapter six). There are again two policy instruments that are examined in this case. The first is the trans-European networks for energy guidelines (TEN-E). These guidelines were revised in 2003 and in 2006 (Decisions 1229/2003/EC and 1364/2006/EC laying down guidelines for trans-European energy

12 Introduction networks). The second policy instrument is the 2009 Regulation on the European Energy Programme for Recovery (EEPR, Regulation (EC) No 663/2009). These policy instruments support energy infrastructure projects in both natural gas and electricity (and in the case of the EEPR Regulation, also offshore wind and CCS projects), both internally and externally to the EU. I focus my analysis on the external natural gas infrastructure projects that are supported by these policy instruments. In this case, low levels of CPI are expected over the time 2000-2010, as the promotion of further gas pipelines interrelates in a conflictual manner with long-term climate policy objectives, by locking energy infrastructure into a fossil fuel pathway (see chapter six). Thus, the continued promotion of these policy measures runs counter to objectives to advance CPI.

Expected Policy Expected results: development Case Category instrument dependent variable from 2000 to 2010? 2001 RES-E Medium Directive RE 1 Yes 2009 RE High Directive 2002 EPBD Low-medium EPB 2 2010 EPBD Yes Medium recast 2003, 2006 Yes: in terms of TEN-E Low policy External gas Decisions instruments; no 3 infrastructure in terms of 2010 EEPR Low dependent Regulation variable Table 1.2: Summary of case studies, showing variation on the dependent variable as a main case selection criterion.

The case study selection thus aims to present the greatest variation on the dependent variable possible within the gamut of EU energy policies (while remaining limited) by choosing one case from each of the three categories in the loose categorisation mentioned above. I follow a most-similar case selection strategy, with variation on the dependent variable as a main criterion for selection (Gerring, 2007, 2008). Each of the three cases is expected to demonstrate differing levels of CPI overall in their policy outputs and processes (low, medium, high). Cases were chosen also on the basis that

13 Introduction they evolved over the course of 2000 and 2010 (with, in two cases, a corresponding expected evolution in the level of CPI to be found). Thus, I may find that at least some of the results of the case study analysis can be generalised to EU energy policy as a whole. However, given the limited number of case studies, it may be difficult to generalise the results to overall EU energy policy, and further empirical research is likely to be required to test these results against cases of CPI in other policy sectors.

1.3.2 Case study methods and techniques The case studies are analysed by process tracing (Bennett, 2008; Checkel, 2008), using three main techniques: document analysis, literature review and semi-structured interviews with individuals involved in policy development.

Process tracing involves “looking at evidence within an individual case” (Bennett, 2008, p. 704). It means providing an historical narrative of a case that produces multiple observations or pieces of evidence that can outline a causal chain linking the independent and dependent variables (Checkel, 2008, p. 115). The data used for process tracing is usually qualitative in nature (ibid., p. 116), but in this research project I drew on data from various sources (including some quantitative data). These include academic literature and texts, databases of climate and energy statistics, policy documents, media reports and interview data.

For each of the case studies, I carried out a qualitative assessment of the documents in the policy process leading to the adoption of the policy measure. This includes documents such as:

• Policy strategy papers (such as green and white papers) published by the Commission; • The Commission’s policy proposal and (where applicable) its impact assessment; • Reports on the consultation process carried out by the Commission, including submitted written opinions from stakeholders (where applicable); • Committee reports in the Parliament; • Adopted amendments in the Parliament; • Council and European Council conclusions; • Council common positions on specific policies;

14 Introduction

• Responses from the Commission (if any) and amended proposals (if applicable); • Final legislative act.

In addition, for each case study, I examined, where available, position papers and opinions of external pro-climate stakeholders, as well as media reports. Media reports are particularly useful as a tool to understand the informal negotiations in the process of a particular policy instrument. Academic literature on the empirical case study was also drawn upon to complement the data collected through documents and media reports. Reports and databases of energy and climate statistics provided quantitative data on the trends in each of the policy sectors (such as the annual Statistical review of world energy from British Petroleum, IEA statistics, Eurostat statistics and reports and statistics of the EEA). Finally, I carried out a number of interviews to complement the data already collected.

Interviews served a complementary data collection function in this project (Hopf, 2004, p. 203). The interviews were carried out for the purpose of confirming/challenging information on the level of CPI in the policy process – namely on the involvement of internal and external pro-climate stakeholders in the policy processes in each of the case studies, and the (perceived and articulated) recognition of the functional interrelations of climate policy objectives in the process (see chapter three and Table 3.1). In total, I carried out 26 interviews with EU energy policymakers and stakeholders, including representatives from the Commission, the Parliament, and from environmental NGOs, industry associations, companies, research institutes and think tanks, and one interview with a member of a national parliament (see Annex III for a list of interviews). The interviews therefore were semi-structured with some guiding questions prepared in advance that allowed enough flexibility for a dialogue to develop (Hermanns, 2004; Hopf, 2004; Rathbun, 2008). The interviews were carried out on the basis of anonymity, and respondents spoke of their personal experience within the context of their organisations.

1.4 Structure and main findings of the dissertation This PhD dissertation is divided into seven chapters. Following this introductory chapter, chapter two discusses the context and background within which the later cases are analysed. This includes a discussion of some of the main concepts used

15 Introduction throughout this work. I briefly outline what I mean by “policy”. Given the EU focus of this research, I describe the EU policymaking process, and the main institutional actors in the process. Next, I focus on the historical context of climate and energy policy in the EU, including the external context that may have affected developments of climate and energy policy in the EU. In sum, chapter two provides the broader context for the analysis of the three case studies in chapters four to six.

In chapter three, I focus on the conceptual framework that informs the study. This includes, first, a conceptualisation of “climate policy integration”, drawing on literature on policy coherence, coordination and integration, and, especially, environmental policy integration. CPI is considered here as a matter of degree (or of varying levels) in both the policy process and output. CPI means including or integrating climate policy objectives into the policy development and policy output of another policy measure. From this discussion, I outline how CPI can be operationalised for an analysis in EU policy. I present benchmarks for measuring levels of CPI in both the policy process and the policy output, with indicators to situate the levels found on a fivefold scale from very low, to very high, over low, medium and high (see Tables 3.1 and 3.2). Next, I focus on the development of the explanatory framework, and identify four main explanatory variables to be applied to the analysis in each of the case studies. These variables are: functional interrelations, political commitment, institutional and policy context, and the process dimension. The first three of these explanatory variables can help explain the level of CPI in both the policy process and policy output, and the fourth factor can help explain CPI in the policy output only (see also above). These four variables were inspired by and derived from the myriad of variables used in the EPI literature, combined with a review of explanatory variables highlighted in general theories of European integration. As I explain in chapter three, I do not pursue a theory-testing exercise, but rather use theory to help develop a comprehensive, yet manageable, conceptual framework for a study of CPI in EU policymaking.

In chapter four, I discuss the first case study – EU RE policy. RE policy in the EU has developed during the 2000s from the 2001 RES-E Directive (2001/77/EC) to the 2009 RE Directive (2009/28/EC). There are, thus, two sub-cases under examination in chapter four. The chapter begins with a description of early developments in the EU on RE policy, followed by a description of the benchmarks against which the levels of

16 Introduction

CPI in the policy output and policy process of EU RE policy are measured. These benchmarks are outlined in reference to the very high levels of CPI that could be expected in this policy area. Establishing these very high levels, for this and for each of the case studies, involves assessing the role played by (in this case) RE on the road to decarbonisation by 2050.9 With reference to a wide selection of studies on the potential road to decarbonisation by 2050, I can derive a scenario of very high levels of CPI in the policy output (some of the scenarios examined in chapters four to six are: EREC, 2010; EREC & Greenpeace, 2010; European Commission, 2011a; EWEA, 2011; ExxonMobil, 2012; Heaps, Erickson, Kartha, & Kemp-Benedict, 2009; Shell International BV, 2008; WWF, 2011, among others). Next, I analyse the levels of CPI in the policy processes and outputs of both the 2001 RES-E Directive and the 2009 RE Directive. In this case, the level of CPI in the 2001 RES-E Directive was low for the policy output and low to medium for the policy process. In 2009, the level of CPI in the policy output was low to medium but the level of CPI in the policy process increased to medium to high. I explain these levels of CPI, and the changes in the results between 2001 and 2009, by applying the explanatory framework described in chapter three. I conclude for this case study that the level of political commitment is one crucial variable for understanding the level of CPI in the 2001 and 2009 Directives’ policy processes and outputs. In neither case, however, is the level of political commitment sufficient for understanding the levels of CPI found. In the 2001 case, the medium and medium to high levels of political commitment were insufficient to advance CPI in the policy process. In turn, the low to medium levels of CPI in the policy process, combined with insufficient levels of political commitment, resulted in low to medium levels of CPI in the policy output. Similarly, the high levels of political commitment at the time of the 2009 RE Directive may have helped push CPI to the medium to high levels found in the policy process in this time phase. This did not seem to affect the low to medium levels of CPI in the policy output, however, which seems to be related rather to the process dimension and the influence of the Council as a less-committed actor that did not sufficiently recognise the long-term climate objectives throughout the policy process leading to the adoption of the 2009 RE

9 “Decarbonisation” is a term often used to describe the objective of the EU to reduce its GHG emissions by 80 to 95 per cent by 2050. This objective requires such a reduction in emissions that it is expected to transform society into one where there are zero or almost zero carbon dioxide emissions (hence: “decarbonisation”). This term is used occasionally in later chapters of this dissertation.

17 Introduction

Directive (see Table 4.3). Thus, a second crucial variable for understanding the level of CPI in EU RE policy, and the change over time, is the involvement of internal and external pro-climate stakeholders in the policy process – which is an important but insufficient sole variable to advance CPI.

In chapter five, I move to the second case study: an analysis of CPI in the EU’s policy on the energy performance of buildings (EPB). EPB policy is part of the EU’s overarching policy to improve energy efficiency, and aims to reduce the energy consumption of buildings by putting in place measures to improve the energy performance of the building’s envelope, of building equipment and of the heating and cooling systems within a building. In chapter five, I discuss the development of EU policy on energy efficiency over time, and describe some of the challenges in making policy on EPB. The empirical analysis of chapter five focuses on two policy instruments between 2000 and 2010, namely the 2002 EPB Directive (2002/91/EC) and its 2010 recast (2010/31/EU). I outline the benchmarks against which the levels of CPI in the policy process and output for the 2002 and 2010 EPB Directives are measured, with reference to a selection of scenarios on the role of buildings in decarbonisation to 2050 (BPIE, 2011; European Commission, 2011a; Heaps, et al., 2009, among others). In this case, the level of CPI in the policy process of the 2002 EPB Directive was found to be low to medium and low in the policy output (see Table 5.2). Interestingly, these levels did not improve to 2010 (although there were some slight increases for CPI in individual indicators in the policy process leading to the 2010 recast Directive). I apply the explanatory framework to these results in an attempt to understand the lack of improvement in the levels of CPI found between 2002 and 2010. Interestingly, in this case, the level of political commitment to combating climate change generally increased over the same time period without resulting in similar increases in the levels of CPI. Additionally, the higher level of political commitment to combating climate change did not translate into a similarly higher level of political commitment to advancing CPI in the EPB sector. The other changes over the time period included a slight increase in the involvement of external pro-climate stakeholders in the policy process (one of the indicators for CPI in the process). In both the 2002 and 2010 Directives, levels of CPI in the policy process were too low to advance CPI in the policy output. This case study seems thus to highlight again that political commitment, while important, is an insufficient criterion

18 Introduction for advancing CPI, but also highlights the role of the Council in the policymaking process. In the EPB case, the Council was the only internal or external actor that seemed reticent to advance on CPI, and managed to ensure the policy remained weak (see Table 5.3).

In chapter six, I analyse the third and final case study in this research project, namely the external gas infrastructure policy of the EU, focusing on the time period from 2000 to 2010. During this decade two main policy instruments were agreed in the EU to support the development of natural gas infrastructure to import gas into the EU: the TEN-E guidelines (Decisions 1229/2003/EC and 1364/2006/EC) and the 2009 EEPR Regulation (Regulation (EC) No 663/2009). These policy instruments aim to promote energy infrastructure projects within the EU and across the EU’s borders (external infrastructure) – natural gas importing infrastructure is only one part of these policy instruments. In chapter six, I describe the natural gas story of the EU, including EU production, consumption, imports and infrastructure for imports. Against this background, and in reference to the discussion in chapter three and various scenarios to 2050 (see references listed above), I describe what very high levels of CPI in the policy process and output of these policy instruments for external gas import infrastructure would look like. I then proceed to analyse the levels of CPI in the policy process and output of the external gas infrastructure policy under the 2003 and 2006 TEN-E guidelines and the 2009 EEPR Regulation. In this case, policies to promote external gas infrastructure display no real evidence of CPI. For both the TEN-E guidelines and the EEPR Regulation the levels of CPI found in the policy outputs and processes are none or very low. How can such levels of CPI be explained? It became clear in this case that political commitment to combating climate change could not advance CPI in sectors where policymakers do not recognise the nature of the functional interrelations between their sector and long- term climate policy objectives. With policies promoting external gas infrastructure likely (but not guaranteed) to lock EU energy infrastructure into a fossil-fuel system for many years, these interact in a conflictual manner with long-term climate policy objectives to decarbonise the EU. The conflictual and rather more indirect (or distant) functional interrelations between external gas infrastructure policies and long-term climate policy objectives may mean that unless explicitly articulated, these interrelations do not enter the policy discussion. This case study seems to complement

19 Introduction the previous two in highlighting the functional interrelations of the policy sectors as one of the crucial factors that has to complement high levels of political commitment, both to combating climate change and to advancing CPI, before high levels of CPI can be expected. The lack of pro-climate stakeholders involved in this case also points to another finding: that general “openness” to the involvement of external stakeholders is not a guarantee that they will be part of the policy process and will push CPI.

Finally, in chapter seven, I conclude the research by drawing out the broader findings of the work. I begin with a reminder of the research question that motivated the research project and assess how the case study analysis succeeded in providing answers to the question. I highlight the main findings and compare these results across cases. A number of main conclusions (on both the case results and on the explanatory framework) include:

• The level of CPI in the EU’s RE, EPB and external gas infrastructure policy is found to be insufficient to achieve long-term climate policy objectives to 2050; • The level of CPI among these three energy policies of the EU varies according to policy measures and has barely advanced between 2000 and 2010; • Although changes in policy have occurred between 2000 and 2010, these changes are often the result of catch-up governance – where the BAU scenario shifts as past policies fail, and new policies may include incremental improvements for the climate, yet have less time in which to achieve their objectives; • In the three cases, the lack of a long-term policy perspective may prevent meaningful levels of CPI emerging; • The recognition and consideration of the (nature of the) functional interrelations between the policy being developed and long-term climate policy objectives in the policy process seems to be a crucial first-order variable for the advancement of CPI; • Political commitment, both to combating climate change and to advancing CPI, is important, but from the analysis here it appears to be insufficient on its own for explaining (or advancing) CPI;

20 Introduction

• The institutional and policy contexts within which policy is developed seem to provide a background context that can add nuance to explanation, rather than serve as a crucial explanatory variable; • The process dimension was found to play a role in understanding the level of CPI in the policy output, yet it appears that this explanatory variable needs to be broken down into its various elements for a clearer understanding.

Additionally, it is clear that no single explanatory factor could adequately explain, in isolation, the levels of CPI found in the policy process and output of the case studies. This research project thus contributes to advancing empirical and conceptual understanding of CPI in the EU. The conceptual framework developed and applied in this research project could benefit from further testing and empirical work, perhaps in sectors such as transport or industrial policies.

Furthermore, this research project highlighted a number of broader questions that are prompted by the results and conclusions found here. The relationship between CPI and EPI is one that is only at the beginning of its conceptualisation and theorisation. Research comparing the empirical realities of these two concepts could help answer more philosophical questions on the value of CPI and EPI. A broader question on the challenges of long-term policymaking in democratic political systems is another issue that could benefit from greater exploration. An analysis of CPI across democratic systems, with 2050 policy objectives as operationalisation benchmarks, could help shed light on the opportunities and challenges for advancing CPI and/or for combating climate change.

In sum, this research project discovers varying levels of CPI among three cases of EU energy policy. I explain this variation with reference to four explanatory variables: functional interrelations, political commitment, institutional and policy context (explaining levels of CPI in both the policy process and output) and the process dimension (explaining levels of CPI in the policy output only). Of these four variables, the (recognition of the) nature of the functional interrelations appears to be a crucial first-order variable for any demand for CPI. Combined with, especially, the level of political commitment both to combating climate change and to advancing CPI, a fuller understanding of the levels of CPI is already found. The remaining two factors can reinforce the explanatory value of these two primary factors, or else serve to provide

21 Introduction the nuance of the background context. The process dimension, particularly, may need to be broken down into its various elements before its explanatory value becomes clear.

This project therefore contributes to advancing knowledge within academic literature on EPI and CPI, especially in the EU. While the three case studies provide rich empirical data and results, the operationalisation and analysis of CPI for both the policy output and the policy process represent a framework that can be replicated in other research projects. The explanatory framework contributes to the development of an encompassing, yet manageable, framework for future research. I bring together various strands and studies on EPI and CPI into a single frame for application to studies on the EU level. Such a study could provide a basis for further empirical and conceptual development in future research studies.

Additionally, the research project highlights some points of note for EU policymakers in the epilogue. Simply stating political commitment to combating climate change seems to be an insufficient act for concrete action on CPI to follow. This political commitment must be backed up with policy measures, but by itself cannot guarantee CPI. Long-term climate policy objectives require explicit consideration in policymaking – a challenge for politicians and policymakers alike. An explicit effort by policymakers to assess sincerely the functional interrelations between long-term climate policy objectives and sectoral policy measures is crucial for meaningful levels of CPI (and procedures may be applied to ensure this takes place). This recognition of functional interrelations can be provided by a high level of involvement of pro- climate stakeholders in the policy process, but this is not a guarantee by itself. Procedural measures could also be adopted to ensure that these functional interrelations (even, or especially, when they are indirect and less obvious) are explicitly part of the policy discussions. Policymakers may need to consider trade-offs among different policy objectives to achieve decarbonisation (possibly also trade-offs among the three objectives of energy policy – competitiveness, sustainability and security). Finally, informal coordination and interaction among policymakers may need to become more institutionalised to ensure CPI occurs across policy sectors.

Context and background

Before embarking on a study of climate policy integration into the EU’s energy policy, it is necessary to situate the study in the broader context of EU climate and energy policy development over time. In this chapter, I discuss EU policymaking generally and the historical development of EU energy and climate policy, including in relation to international events and developments. This chapter sets the scene in advance of the conceptual discussion in chapter three.

When it comes to combating climate change, the EU has long aimed for global leadership (Bretherton & Vogler, 2006; Gupta & Grubb, 2000; Oberthür, 2007; Oberthür & Roche Kelly, 2008; Schreurs & Tiberghien, 2007). Yet developments on the international stage have not always lived up to the notion of the EU leading and others following – as was made plain in the fifteenth Conference of the Parties (COP) to the United Nations Framework Convention on Climate Change (UNFCCC) in Copenhagen in December 2009. At this COP, EU negotiators were side-lined while final agreements were made behind closed doors among a small group of national leaders (Dimitrov, 2010; Falkner, Stephan, & Vogler, 2010; Oberthür, 2011).

Credible leadership requires action to be taken on the domestic level to demonstrate an ability and willingness to move forward, and to prove to the international world that, in the EU, action follows from rhetoric (Oberthür & Roche Kelly, 2008; Wurzel & Connelly, 2011b). Domestic/internal EU climate policy measures are thus necessary for credible international leadership. Action within the EU to combat climate change is necessary, not only to live up to these international expectations, but also to preserve the EU’s own self-interest, solidarity, and even survival (Behrens & Egenhofer, 2011, p. 219).

The process to agree on internal policy and legislation in the EU is complex, with many actors, institutions, preferences and interests to consider. Additionally, the climate issue is one that crosses traditional sectoral policymaking processes and has proven challenging for policymakers to respond to comprehensively (Jordan, Huitema,

23 Context and background

& van Asselt, 2010). The EU, under such circumstances, has nevertheless managed to garner the political and the institutional support to introduce several pieces of legislation aimed at combating climate change over the course of its history of climate policy.

This chapter opens with a discussion of the policymaking process in the EU, and follows with an historical overview of climate and energy policy development in the EU in three broad time phases – the 1980s and earlier; the 1990s; and the 2000s. I chose this division of time phases for two reasons. First, while climate policy in the EU over time has often been linked with major landmarks in the international negotiations (such as the 1992 signing of the UNFCCC, the 1997 signing of the Kyoto Protocol, and the preparation in the 2000s for the post-Kyoto timeframe), the same cannot be said of the many developments in EU energy policy. What becomes clear in the discussion below is an acceleration of policy development in both policy fields especially in the 2000s. Second, the empirical analysis of this project focuses on this first decade of climate and energy policy development of the twenty-first century (from 2000 to 2010). I therefore treat the developments of this decade together and follow the same pattern in discussing earlier developments. Throughout the discussion on the development of EU climate and energy policy, I refer to the international developments in climate policy (along with other external events relevant to the development of energy and climate policy) that, at times, drove (or blocked) EU internal policymaking. What becomes clear by the end of the chapter is that EU climate and energy policies have become more closely linked over time, with some energy policy measures being touted as responses to climate change. With about 80 per cent of GHG emissions in the EU coming from the energy sector (EEA, 2010, p. 31), such motivations for energy policy development seems logical, and this has helped advance EU energy policy also.

2.1 Concepts and definitions It is worth providing some explanation of the concepts used throughout this project, as well as providing an historical overview of the development of climate and energy policies in the EU.10 I assess and explain the extent of climate policy integration (CPI)

10 Although, especially in this chapter, I discuss the history of the European Union when it was called the European Economic Community, or the European Community, etc., I use the term “EU” throughout for ease of understanding and for consistency.

24 Context and background into certain EU energy policies, with a focus on the policy process and the policy output of these cases of energy policy. I thus do not focus the assessment on, e.g., the effectiveness of a policy measure’s implementation. In this section, therefore, I begin with a general discussion of the policy cycle and the policymaking process in the EU.

2.1.1 What is policy? The Oxford English Dictionary defines “policy” as “a course or principle of action adopted or proposed by an organisation or individual”. Reaching agreement on a policy measure, and on the tools to achieve its goals, takes place among decision- making parties during the policymaking process – in the case of the EU’s co-decision procedure,11 among the European Commission (“the Commission”), the European Parliament (“the Parliament”) and the Council of the European Union (“the Council”; see section 2.1.3 below for more details).

Scholars of policy analysis have long tried to simplify the complex structure of policymaking into a digestible and accessible form. Kingdon (2003), in his influential study on agenda-setting in US policymaking, provides a definition of policymaking as a set of processes “including at least (1) the setting of the agenda, (2) the specification of alternatives from which a choice is to be made, (3) an authoritative choice among those specified alternatives, and (4) the implementation of the decision” (Kingdon, 2003, pp. 2-3). The most common form of summarising the policymaking process overall has been the use of the “policy cycle” tool to identify the different stages in the process. The policy cycle tool aims to conceptualise policymaking as a continuous loop where each stage in the policymaking process flows smoothly to the next and feeds back to the beginning of the cycle again (see, e.g., Figure 2.1). It is a tool that has been criticised for its over-simplification of a complex reality (Crabbé & Leroy, 2008). In response to these criticisms, the identification of the steps in the policy cycle has developed over time, and it is now acknowledged that the various steps often overlap with each other, therefore taking place more or less consecutively or concurrently (Crabbé & Leroy, 2008, p. 3). While the policy cycle may not be a

11 The co-decision procedure was renamed the “ordinary legislative procedure” in the Lisbon Treaty (the Treaty on the Functioning of the European Union). I continue to use the term “co-decision procedure” throughout this work, however, as this was the name of the procedure for the majority of the time under examination in the empirical case studies (between 2000 and 2010).

25 Context and background perfect tool for policy analysis, it remains a useful broad outline of the various parts of the policy process overall, and can also be valid for the EU level.

Figure 2.1: The policy cycle.

Source: Adapted from Crabbé & Leroy, 2008 (p. 3).

2.1.2 Policymaking in the EU While Figure 2.1 outlines several stages in the policy cycle, this research project focuses on analysing the policy process (which links broadly to the “policymaking”, “selecting options” and “forming opinions” steps of Figure 2.1) and policy output (which is loosely comparable to the “political decisions” step in Figure 2.1) at the EU level. Policymaking in the EU is often considered a complex process involving many interests, actors and preferences, and there exists a correspondingly numerous proliferation of theories to explain the European integration and EU policymaking processes (Hauser, 2011; Lelieveldt & Princen, 2011; Wallace, Pollack, & Young, 2010).

Confusion may also exist about what is the “EU”, what is its nature, and therefore what is the background context for discussing EU policymaking. The very nature of the EU has been debated at length by many scholars over the years: is it an international organisation (Pollack & Hafer-Burton, 2010; Tsebelis & Garrett, 2001)? Is it an “actor” (Bretherton & Vogler, 2006)? Is it the realisation of a “cosmopolitan

26 Context and background vision” (Beck, 2009)? Additionally, the EU as a single entity has been differently described as a “normative power” (Manners, 2002, 2006) and as a “market power” (Damro, 2012). I do not discuss the nature of the EU nor its power in the world, but I rather focus on internal EU policymaking processes and outputs in energy policy. Therefore, in this project, the “EU” is understood as the supranational policymaking level involving (predominantly, but not exclusively) the three co-deciding institutions: the Commission, the Parliament and the Council. However, these institutions do not make policy in a vacuum – they interact with the politically important European Council (see section 2.1.2.4 below explaining the European Council) and with many “interest groups” that lobby the policymaking institutions (and notwithstanding the wider context of international events or developments that may also raise or lower policy priorities).

There are several different methods or procedures of policymaking in the EU, including (but not limited to): the co-decision procedure (which was renamed the ordinary legislative procedure in the Lisbon Treaty), the intergovernmental method, and the open method of coordination. The intergovernmental method puts member states at the centre of decision making, with the Commission sharing the right of initiative, and the Parliament playing, at best, an advisory role. The open method of coordination does not in itself lead to binding legislation but aims to promote policy coordination and learning among member states (Lelieveldt & Princen, 2011), which can potentially lead to policy development in future. For the purposes of this research project, I focus on the co-decision procedure, as, in each of the three cases I examine in later chapters, policy measures were agreed under these rules (most instruments studied were agreed before the entry into force of the Lisbon Treaty when the co- decision procedure became the ordinary legislative procedure).

The co-decision procedure extends, in the Lisbon Treaty era (post-2009), to the majority of policy sectors (more than 95 per cent of Community legislation; Lelieveldt & Princen, 2011, p. 67). Under this procedure, the Commission holds the sole right of initiative to make legislative proposals. Once the Commission publishes a policy proposal, it moves through negotiations between the Parliament and Council, with both these institutions enjoying full rights to amend and agree on the policy. Parliament adopts amendments by simple majority, with qualified majority voting (QMV) as the final decision-making rule in the Council (Art. 294 TFEU; see section

27 Context and background

2.1.2.2). Below, I discuss more specifically the role of each of the three institutions involved directly in the EU’s co-decision procedure, and how they are relevant for the purposes of this research project on CPI into the EU’s energy policy.

2.1.2.1 The European Commission Under the co-decision procedure, the Commission is the sole initiator of policy proposals, giving it a central role in the policymaking process of the EU.12 It shares executive power with the Council (see section 2.1.2.2). The other main tasks of the Commission include overseeing the implementation of policies, and some external relations (Barnes, 2011, pp. 41-42). The Commission is led by a college of commissioners, including the President of the Commission, comprised of one commissioner per member state, each of whom is supposed to work independently and in the interests of the EU as a whole (Art. 245 TFEU).13 Each Wednesday, the college of commissioners meets and decides on the agenda items (as proposed by the Commission President), with decisions being taken either by consensus or majority (Commission Rules of Procedure, (2010/138/EU), Arts. 4, 8), although consensus is the preferred option (Egeberg, 2010, p. 129).

The Commission is involved throughout the policy process, from initiation to agreement. It interacts with the Parliament and Council throughout the negotiations and while overseeing implementation. It plays a political as well as administrative role. As the EU has enlarged, the Commission has grown with it, and has developed its political skills in building consensus among 27 commissioners (Barnes, 2011, p. 43). Nevertheless, its ability to shape decisions in the EU is limited (and has been limited) by its legal competence, expertise and size. In climate and energy policymaking, the Commission has moved from few, but radical, policy proposals in the early 1990s (such as the failed CO2/energy tax) to new policy proposals in the late 1990s, to seizing opportunities for policy development that arose due to insufficiently ambitious past policies and external events (such as in the case of the reformed ETS Directive 2009/29/EC negotiated in 2008; see section 2.2 below for more detail).

12 With the TFEU, a new “citizens’ initiative” allows for direct pressure from citizens of the EU on the Commission to initiate policy proposals on a certain issue (Art. 24, TFEU). The rules for citizens’ initiatives are elaborated in Regulation (EU) No 211/2011 on the citizens’ initiatives.

13 Article 245 of the TFEU states: “the Members of the Commission shall refrain from any action incompatible with their duties. Member states shall respect their independence and shall not seek to influence them in the performance of their tasks”.

28 Context and background

The President of the Commission has often played a crucial role in providing political commitment to pushing or blocking climate and energy policy. President Jacques Delors (Commission President 1986-1994), for example, oversaw the early days of Commission policy development on climate change (while not always agreeing with the policy ideas of his Environment Commissioner; Barnes, 2011; Skjærseth, 1994). Presidents Jacques Santer (1995-1998) and Romano Prodi (1999-2004) are considered to have been weak Commission Presidents, both in terms of managing their Commission and in showing commitment to the climate issue (Barnes, 2011). President José Manuel Barroso (since 2004) has been considered to come late to the climate issue, with his policy preferences in the early days of his first mandate residing in jobs and growth and ignoring climate change and environmental concerns generally (ibid.). He later provided more political backing to taking action on climate change (from about 2007), due to energy security issues rising on the agenda, the recognition of climate change as an issue of concern to European citizens (Eurobarometer, 2008, 2011), the opportunity provided by the climate problem to push European integration, and the advancing climate-leadership ambitions of the EU as a whole (Barnes, 2011; Oberthür & Roche Kelly, 2008). In its evaluation of the first Barroso Commission (2004-2009), Euractiv highlights that “most observers agree that the EU’s historic agreement to cut greenhouse gas emissions by 20% by 2020 and the decision to put in place the necessary steps to make it happen has been the greatest success of the Barroso Commission”.14 In his second mandate (2009- 2014), Barroso reshuffled his Commission portfolios to include a Directorate-General (DG) for Climate Action from 2010, headed by Commissioner Connie Hedegaard (2010-2014).15

2.1.2.2 The Council of the European Union The Council (also known as the Council of ministers) is the institution that brings together representatives of member states at the ministerial level. It operates in different sector-specific configurations. The Council formations most relevant for this research project include the transport, telecommunications and energy Council, which meets about six times a year, and the environment Council, which meets about four

14 See http://www.euractiv.com/priorities/2004-2009-evaluation-barroso-com-linksdossier-188516, accessed: 19 February 2013.

15 See http://ec.europa.eu/dgs/clima/mission/index_en.htm, accessed: 19 February 2013.

29 Context and background times a year. The Council’s role in policymaking includes acting as co-legislator with the Parliament. It can also, through the adoption of conclusions and recommendations, help push policy issues up or down the agenda (Lelieveldt & Princen, 2011). But as one of three co-legislating institutions, the Council cannot easily move forward on policy on its own, although it can block policy developments (Oberthür & Dupont, 2011).

The decision-making process in the Council under the co-decision procedure is, and was, qualified majority voting (QMV).16 Decisions need a high level of support in the Council for QMV. The number of votes per member state is weighted according to population, with no member state holding more than 29 (Germany, France, UK and Italy) or less than three votes (Malta; Corbett, Peterson, & Bomberg, 2012, pp. 58-59). From 2014, following the entry into force of the Lisbon Treaty in 2009, a qualified majority requires 55 per cent of the member states making up 65 per cent of the population of the EU, with a blocking minority required to represent more than 35 per cent of the population (TFEU, Art. 238(3)).17

The details of policy proposals are first negotiated at the level of working groups in the Council, before negotiations move to the Committee of Permanent Representatives (COREPER). The Permanent Representatives are the heads of the delegations of member states to the EU, and they play a “critical” role in EU policymaking (McCormick, 2005, p. 90). As of 2012, there were close to 200 working groups in the Council (Council of the European Union, 2012). In COREPER I, some of the more technical issues of a specific policy proposal are discussed among the member states’ deputy permanent representatives. In COREPER II, the permanent representatives themselves meet to discuss the final political issues in the policy proposal, although not all policy proposals would be discussed in both COREPER I and II. With this elaborate system of preparation on a policy proposal, the Council meetings themselves do not discuss many policy proposals in-depth. A-point items on the Council meeting agenda are items that do not require further discussion after the preparatory meetings, and need simply the accord of the Council. B-point agenda

16 Where the Council votes to amend a proposal from the Commission, the unanimity rule applies except in certain circumstances (Art. 293, TFEU, see Annex II).

17 Until 2014, the Lisbon Treaty accounts for a triple majority for QMV: 51 per cent of member states, 62 per cent of the EU’s population and a majority of votes (Corbett, Peterson, & Bomberg, 2012, p. 59).

30 Context and background items, however, call for further discussion and agreement at the ministerial level before they can be adopted. In reality, the majority of Council decisions fall under A- point agenda items, meaning the negotiations have taken place outside of the Council meeting itself (Lelieveldt & Princen, 2011, p. 61, suggest that up to 80 per cent of Council decisions are A-point agenda-items).

The Presidency of the Council rotates every six months among the member states, and provides member states with the opportunity to steer the policy agenda in line with their own priorities (Warntjen, 2007). Although the Presidency’s room for manoeuvre may be limited, due to the inheritance of policy items already on the agenda and external events that require action, the Presidency can nevertheless play a facilitative role in pushing or blocking the adoption of policies (Lelieveldt & Princen, 2011, p. 60). In climate and energy policy, the Presidency could demonstrate political commitment to the climate issue by pushing forward on policies during its six-month mandate. This was the case with the Dutch Presidency in 1997, for example, under whose mandate the EU proposed to reduce GHG emissions by 15 per cent in time for the international climate negotiations in Kyoto, and with the French Presidency in 2008 that pushed for agreement on the climate and energy package under its mandate (before the less pro-climate policy Czech Presidency took over in 2009; Burns & Carter, 2011, p. 64; Oberthür & Dupont, 2011).

In climate and energy policymaking, the Council has historically often blocked or weakened the Commission’s policy proposals. The Commission’s flagship policy proposal in the 1990s on a CO2/energy tax (which required unanimous support in the Council) was blocked by the Council and eventually abandoned (European Commission, 2001a; Skjærseth, 1994; Wettestad, 2000). Other policies agreed during the 1990s were weakened significantly by the Council, such as the SAVE programme for energy savings and the ALTENER programme on renewable energy (see more in section 2.2.2 below, and in chapters four and five; Oberthür & Pallemaerts, 2010a). As time went on, however, the Council began to demonstrate more political commitment to furthering climate policy and to energy policy development in the EU. Motivated in part by ambitions to demonstrate leadership internationally, the Council agreed to more far-reaching and binding policy developments to combat climate change internally in the 2000s (Oberthür & Dupont, 2011; Oberthür & Roche Kelly, 2008; Wurzel & Connelly, 2011a).

31 Context and background

2.1.2.3 The European Parliament As the only directly elected institution of the EU, the Parliament represents the European citizen in the policymaking process. It therefore is considered the most suited institution for voicing citizen concerns (Tanasescu, 2009, p. 48). Its role in policymaking has evolved with the Treaties over time so that it has gained formally equal power with the Council as a co-legislator in policies under the co-decision procedure – the most commonly used legislative procedure in EU policymaking (Lelieveldt & Princen, 2011).

Members of the European Parliament (MEPs) are organised along ideological (rather than national) lines in political groups.18 Much like in the Council, preparatory work on a policy proposal takes place outside the plenary of the Parliament. Parliamentary committees, composed of a number of MEPs from different political groups, prepare and debate the proposals before final decision-making takes place in the plenary sessions. The committee submits a report to the plenary that may (or usually does) contain proposed amendments to the Commission’s policy proposal. In the plenary session, MEPs vote on each of the proposed amendments (Corbett, Jacobs, & Shackleton, 2007). As of 2012, there were about 22 standing parliamentary committees, with the two committees most relevant for this research project being the industry, research and energy committee (ITRE) and the environment, public health and food safety committee (ENVI).

In addition to co-legislating activities, the Parliament can influence the policymaking agenda to a certain extent by adopting its own reports and resolutions. In this way, it can try to influence the Commission to move forward with a policy proposal on a certain issue (Lelieveldt & Princen, 2011, p. 67). However, as with the Council, the Commission is not obliged to follow the urges of the Parliament and retains the sole right of policy initiation.

18 The seven political groups (in 2013) include: European People’s Party (Christian Democrats, EPP); Progressive Alliance of Socialists and Democrats in the European Parliament (S&D); Alliance of Liberals and Democrats for Europe (ALDE); Greens/European Free Alliance; European Conservatives and Reformists (ECR); European United Left – Nordic Green Left (GUE/NGL); and, Europe of Freedom and Democracy Group (EFD).

32 Context and background

Traditionally, the Parliament is considered the most environmentally friendly of the three co-legislating institutions of the EU (Burns & Carter, 2011),19 although its ability to push for ambitious legislation is often limited by the continued political weight of the Council (even with QMV in the Council). In climate and energy policy development, the Parliament has often been overshadowed by Council in negotiations, or Council’s reactions have only weakly responded to Parliament’s demands (Burns & Carter, 2011). For example, in the policy developments in the late 2000s on the climate and energy package, commentators remarked that Parliament’s ability to push for ambitious policy was hampered by member state reticence. The role of Parliament in 2008 was rather to ensure that policy was adopted, than to push for more ambition on combating climate change. As Burns and Carter (2011, p. 69) highlighted: “[f]or the Parliament the passage of the climate change and energy package served to underline the limitations of its powers under co-decision when the Council is intransigent” (see section 2.2 below for more on the development of EU climate and energy policy; see also chapter four).

2.1.2.4 The European Council Although the European Council is not one of the three decision-making institutions under the co-decision procedure, it is nevertheless an important political actor in the EU. With the entry into force of the Lisbon Treaty in 2009, the European Council became an official “institution” of the EU (Art. 9), with a President at its head (President Herman van Rompuy, since 2009). Even without this status earlier, the European Council was influential in pushing the political priorities of the EU. The European Council brings together the heads of state and government (Prime Ministers or Presidents) of the member states at summit meetings organised about four times a year (twice per Presidency).20 Its importance as an institution stems from its political weight and its role and ability to provide the impetus for moving forward on specific issues. The “Presidency conclusions” coming from the meetings of the European Council often indicate the issues of political importance for member states at the time

19 Research has been disputing the green credentials of the Parliament in recent years, especially in the aftermath of enlargement in 2004 and 2007. See, e.g. Carter & Burns, 2009, and Rasmussen, 2012, who argue that the Parliament is an “environmental pragmatist” rather than an environmental champion.

20 See: http://www.european-council.europa.eu/the-institution, accessed: 19 February 2013.

33 Context and background and can result in action from the Commission to initiate certain legislative proposals (Oberthür & Dupont, 2011; Tallberg, 2004; Warntjen, 2007).

In climate and energy policymaking, the conclusions of the European Council have over time come to provide the political impetus to develop more internal EU policy and to promote the EU’s climate leadership ambitions internationally (Oberthür & Dupont, 2011). In March 2007, the European Council agreed to an independent commitment to reduce GHG emissions in the EU by 20 per cent by 2020 compared to 1990 levels and called on the Commission to propose legislation (European Council, 2007). Some research has shown that on climate change issues, the political commitment or will of the European Council to act on climate change increased from little interest in the climate issue in the 1990s and early 2000s, to a high level in about 2007-2008 (Oberthür & Dupont, 2011, p. 82).

2.1.2.5 Other actors Beyond the three co-deciding institutions of the EU policymaking process, and the political impetus provided to the process by the European Council, two other institutions are generally consulted under the co-decision procedure. These are the European Economic and Social Committee (EESC) and the Committee of the Regions (COR). The EESC claims that it is “a bridge between Europe and organised civil society”, divided into three groups: the employers’ group, the workers’ group and the “various interests” group (with 344 members). The COR represents the interests of regions and cities in the EU policymaking process (again with 344 members). Both the EESC and the COR are consultative bodies for the EU and their opinions on the proposals of the Commission are forwarded to all three deciding institutions.21 Their opinions hold no legislative weight, however.

In addition to these representative committees, external stakeholders, lobbyists and interest groups attempt to influence policymaking in the EU. Lobbying efforts take place at all stages of the EU policymaking process. The Commission and Parliament, especially, often seek out expert opinion on certain policy areas. These actors are accommodated in EU policymaking both formally (through official consultation procedures) and informally (through lobbying activities, personal relationships, etc.;

21 See: http://www.cor.europa.eu and http://www.eesc.europa.eu for more information, accessed: 15 March 2013.

34 Context and background

Hauser, 2011; Tanasescu, 2009; Watson & Shackleton, 2003). External actors can play a significant role especially in advancing or hindering CPI in other policy sectors (see also chapter three).

2.1.3 The EU’s co-decision procedure: step-by-step As I focus in this research project primarily on policy decisions taken under the co- decision procedure, I outline here the main steps in that policymaking procedure.

The Maastricht Treaty (1992) first introduced the co-decision procedure, which has been progressively extended to more policymaking areas in subsequent Treaties (Treaty of Amsterdam in 1997, Treaty of Nice in 2001 and Treaty of Lisbon in 2007). The greatest strides forward for expanding the co-decision procedure came in the earlier Treaties of Maastricht and Amsterdam, with fewer advances in later Treaties. This is explained with the different focus of the later treaties (e.g. coping with enlargement for the Nice Treaty, and replacing the efforts towards a constitutional treaty with the Lisbon Treaty), and that the EU had already reached an elevated level of policymaking (Hix & Høyland, 2011; Piris, 2010, p. 71). With the TFEU, the co- decision procedure became the “ordinary legislative procedure” and is defined under Articles 289 and 294. The main steps in the procedure itself did not change between 2000 and 2010. It begins with the Commission sending its policy proposal to the Parliament and Council for deliberation and negotiation. Both the Council and the Parliament have two opportunities (two “readings”) to discuss and agree on the proposal before the proposal is brought to a “conciliation committee” at the third reading.

In the first reading, Parliament can adopt amendments (prepared by the committee in charge of the dossier, and voted on in plenary) to the Commission’s proposal. The Commission provides an opinion on Parliament’s amendments and can, if it wishes, put forward an amended proposal. Council’s first reading can approve the Parliament’s amendments, in which case the act is adopted, or not approve all the Parliament’s amendments and/or suggest some of its own. In this latter case, the Commission provides an opinion on the Council’s amendments and the proposal moves to the second reading.

35 Context and background

In the second reading, there is a deadline to complete negotiations within three months (extendable by one month). Here the process follows a similar pattern as in the first reading. If Parliament agrees with the amendments of Council, the act is adopted. If Parliament rejects the Council’s position, the act is not adopted. However, Parliament can also suggest new amendments to those of the Council, in which case the Commission provides a new opinion on these amendments and the proposal moves into second reading in the Council also. Again, if Council approves the new amendments put forward by the Parliament, the act is adopted. However, if the Council does not approve these new amendments, the proposal moves into the conciliation stage.

In the conciliation phase, the deadline for completing negotiations stands at six weeks, with the option to add an extra two weeks, after the conciliation committee is first convened. The conciliation committee’s members come in equal number from the Council and the Parliament, with Commission representatives also attending the meetings. If negotiations in conciliation are unsuccessful, the act cannot be adopted. The committee sends its agreement to Council and Parliament for the third reading, and, if the proposal is adopted by both institutions (within the time deadline of six to eight weeks), it enters into law (European Parliament, 2012; Watson & Corbett, 2012, pp. 123-127; TFEU, Art. 294).

In addition, negotiations among the three institutions can occur in the form of “trilogues”. These meetings take place in private and in an ad hoc fashion among representatives of the Parliament, Council and Commission. The aim is to reach a timely agreement on amendments that are acceptable to both the Council and Parliament, and that can be approved by all three institutions. Although trilogues are generally first arranged during the second reading (European Parliament, 2012, p. 19), there are no fixed rules. A growing tendency to favour first reading agreements means that ad hoc informal negotiations are often ongoing from early stages.22 Additionally, the number of trilogue meetings in second reading and conciliation has been increasing over time, also due to the time constraints for agreement under these time phases (European Parliament, 2012, p. 20). The trilogue meetings usually include the rapporteur from the Parliament in charge of preparing the lead committee’s proposed

22 See http://ec.europa.eu/codecision/stepbystep/glossary_en.htm, accessed: 13 December 2012.

36 Context and background amendments (and shadow rapporteurs from other political groups in the Parliament are occasionally present); the chairperson of COREPER I or the of the relevant Council working party (with assistance from the general secretariat of the Council); and representatives from the Commission, with the expert in charge of the dossier usually involved, along with representatives from the Commission’s legal service.

Recent studies on the evolution of decision making in the EU point to an increase in the informal negotiations among the institutions over time, and a corresponding increase in early or first reading agreements (De Clerck-Sachsse & Maciej Kaczynski, 2009). The co-decision procedure, as amended under the Amsterdam Treaty (1997), extended the role of the Parliament in the legislative process, but also led to an increase in the workload of the Parliament and an increase in informal contact between the Parliament and the Council even before the first reading (Burns & Carter, 2010, p. 138). Some argue that this development has led to a less environmentally ambitious Parliament generally (Carter & Burns, 2009; Rasmussen, 2012).

Overall, the co-decision procedure in the EU follows to a certain extent some of the steps generally outlined as part of the “policy cycle” (see Figure 2.1). These steps include: selecting and structuring problems (with the Parliament, Council and European Council influencing the policy agenda as they can, but with the Commission holding sole right of policy initiation); submitting policy proposals (the Commission submits to the Parliament and Council); generating, comparing and selecting options (readings in the Parliament and Council, negotiations in trilogues and informal meetings); and, agreement on a final policy proposal leading to the policy output (the final adopted act).

In this research project, I therefore refer to the “policy process” as including the steps from the development and introduction of the policy proposal to the final agreement among the institutions. The “policy output” refers specifically to the final legislative act, and the policy objectives laid out in that document.

2.2 EU climate and energy policy development Climate change is often said to be one of the greatest and most complex problems facing humankind (Jordan, Huitema, van Asselt, Rayner, & Berkhout, 2010; Wurzel & Connelly, 2011b). The European continent is not immune to the impacts of a

37 Context and background changing climate. Human-induced climate change is caused by the emission of potent and long-lived GHGs into the atmosphere, the emissions of which have grown since pre-industrial times. As the most important GHG, carbon dioxide (CO2) provides the benchmark against which all other GHGs are measured, as CO2-equivalent emissions and concentrations in the atmosphere. The main sources of anthropogenic GHG emissions globally can be found in the burning of fossil fuels (oil, coal, natural gas) for electricity generation, transport, industry and in households. Deforestation, landfilling of waste, use of fluorinated gases 23 and agricultural activities also represent a significant, but smaller, proportion of GHG emissions. In the EU, about 80 per cent of GHG emissions are accounted for by energy production and consumption, including in transport and in electricity and heat generation and consumption in industry and in households (EEA, 2010, p. 31). The annual emissions of CO2 grew globally between 1970 and 2004 by about 80 per cent (IPCC, 2007). As European nations were among the first to undergo an industrial revolution in the late 1700s, they have historically contributed greater amounts of GHG emissions to the atmosphere, and the UNFCCC notes in the preamble that “the largest share of historical and current global emissions of greenhouse gases has originated in developed countries”. Thus, when the UNFCCC was signed in 1992, the Convention stipulated “common but differentiated responsibilities” among the signatories, and that developed country Parties should “take the lead in combating climate change and the adverse effects thereof” (Arts. 3(1) and 4(1)). The picture has somewhat changed in the twenty years since then, with the EU’s share in global GHG emissions declining to about 12 per cent in 2012 from about 18 per cent in 1990,24 as the GHG emissions from large developing countries, such as China in particular, continue to rise (CAIT 2005 data25). Figure 2.2 shows the GHG emissions of the EU from 1990 to 2010.

23 The Kyoto Protocol regulated the emissions of a basket of six GHGs in its first commitment period (2008-2012), including three fluorinated or F-gases: perfluorocarbons (PFCs), hydrofluorocarbons (HFCs) and sulphur hexafluoride (SF6). A fourth gas, nitrogen trifluoride (NF3) was added for the second commitment period of the Kyoto Protocol under the “Doha Amendment” in 2012. The fluorinated gases are potent GHGs that are used mainly in refrigeration, air-conditioners, aerosols, foaming agents and in industrial processes as replacements for ozone-depleting chlorofluorocarbons (CFCs).

24 See www.eea.europa.eu, accessed: 22 January 2012.

25 Climate analysis indicators tool (CAIT), www.wri.org/tools/cait, accessed: 22 January 2012.

38 Context and background

Figure 2.2: GHG emissions in EU15 and EU27, 1990-2010.

Source: European Environment Agency (EEA) GHG emission database, www.eea.europa.eu (see also Oberthür & Pallemaerts, 2010a, p. 32).

As well as being an historical emitter of GHGs due to its early industrial revolution, Europe is expected to face differentiated impacts of the changing climate in the future. According to the European Environment Agency (EEA), the main consequences of climate change that will be seen in Europe include “an increased risk of coastal and river floods, droughts, loss of biodiversity, threats to human health, and damage to economic sectors such as energy, forestry, agriculture, and tourism” (EEA, 2010, p. 38). But these impacts are expected to be felt sooner and more severely in the Mediterranean, northwestern Europe, the Arctic and mountainous areas. Unless action is taken to mitigate climate change, according to the Stern review, greater costs will be involved in adapting to the impacts of climate change later (Stern, 2007).

While it is important to keep these historical, scientific and economic arguments in mind when discussing the development of EU climate policy, they certainly do not provide a complete picture. EU action on climate change is also linked to the development of the European integration project in general, and political ambitions to be a credible and unified actor on the international stage.

39 Context and background

Next, I discuss the developments of EU climate and energy policy over time, broadly dividing the discussion among the three decades of the 1980s (and earlier); the 1990s; and, the 2000s (with this first decade of the twenty-first century being the focus of the empirical research outlined in chapters four to six). For each of these time phases, I discuss not only the internal policy developments on climate and energy issues, but also the international events and negotiations that may have pushed domestic EU policymaking forward.

2.2.1 The early days: 1980s and before While the very foundation of the EU can be traced to links in energy policy across borders, climate has a much more recent history. The European Coal and Steel Community Treaty, signed in 1951, brought France, Germany, Italy, Belgium, Luxembourg and the Netherlands together in what would later expand and become the European Union. In 1957, the creation of Euratom, the European Atomic Energy Community, linked the European nations further, again with energy policy as a pillar of integration. Thus, the European integration project began on the strength of energy connections. Yet, since then, there has been limited competence for the European institutions to make policy on energy issues (Duffield & Birchfield, 2011). Climate change, on the other hand, emerged comparatively late in the development of the EU, but competence on the issue, based often on legal competences in the areas of the environment and the internal market, allowed the EU institutions to move forward in policy development, even shortly after the issue arrived on the political agenda (Jordan & Rayner, 2010). Linking to these same competences in the internal market and environment helped the development of EU energy policy also (see Table 2.1).

40 Context and background

Period Energy Policy Climate Policy External events 1950s 1951: (Climate change regarded as a ECSC, European Coal and Steel scientific issue) Community Treaty 1957: EURATOM, European Atomic Energy Community 1960s 1968: (Climate change regarded as a “First guidelines for a Community scientific issue) energy policy”, COM (68) 1040 1970s 1972: 1978: 1972: “Necessary progress in Community “Proposal for a multiannual United Nations energy policy”, COM (72) 1200 research programme in the field of Conference on 1974: climatology” COM (78) 408 Human Environment “Community energy policy. (Stockholm) Objectives for 1985”, COM 1973: S (74) 1960 First oil crisis “Towards a new energy policy 1979: strategy for the European Second oil crisis S Community”, COM (74) 550 World Climate 1978: S Conference, “Energy objectives for 1990 and S Geneva programmes of member states”, COM (78) 613 1979: “The energy programme of the European Community”, COM (79) 527 1980s 1981: 1986: 1985: “The development of an energy Parliament Resolution on the Villach climate strategy for the Community”, COM “greenhouse effect” conference (81) 540 1988: 1987: 1983: “The greenhouse effect and the Single European Act “Proposal for a Council Community”, COM (88) 656 1988: Regulation establishing S Rhodes Council declaration on IPCC established S specific measures of the environment Community interest relating to S energy strategy”, COM (83) 31 “Community energy strategy: progress and guidelines for S future action”, COM (83) 305 1985: “New community energy objectives”, COM (85) 245 1988: “The internal market for energy”, COM (88) 238 Table 2.1: Some of the main policy developments in EU energy and climate policy, 1950s – 1980s.

Sources: http://europa.eu; Birchfield & Duffield, 2011; Buchan, 2009; Damro, Hardie, & MacKenzie, 2008; Jordan & Rayner, 2010; Oberthür & Pallemaerts, 2010b; Oberthür & Roche Kelly, 2008; Pallemaerts & Williams, 2006; Wurzel & Connelly, 2011b. In 1968, the Commission published a memorandum called “first guidelines for a Community energy policy” (European Commission, 1968) in which it called for a common energy policy to counteract distortions in competition due to differences in

41 Context and background energy costs from one member state to another (European Commission, 1968, para. 4). Already in 1968, the Commission highlighted the importance of an internal energy market, and the need to reduce member state dependence on imported energy. The oil crises in the 1970s (in 1973 and 1979) only heightened the risk of high-energy import dependence, yet even these shocks did not move member states towards more integration on energy policy. At this point, energy policy focused on the objectives of “maintaining a regular, stable supply” of energy to the EU (European Commission, 1972, p. 8), and reducing energy import dependence. Energy efficiency and savings already emerged among the strategies for tackling security of supply issues (European Commission, 1972, 1978). In 1974, the Council agreed to energy policy objectives to 1985 in response to the first oil crisis in the early 1970s. These targets included restricting dependence on imported energy to fifty per cent and stepping up energy savings. By 1978, it was already clear that further efforts would be required to meet these targets (European Commission, 1978).

Even in the face of the oil crises, little concrete progress was made in the 1970s to achieve an EU energy policy, despite Commission endeavours. The 1980s, however, saw a generally heightened environmental awareness, with linkages beginning to be made between energy and environmental policies. This followed both the 1972 United Nations Conference on the Human Environment in Stockholm and the World Climate Conference in Geneva in 1979 (Pallemaerts & Williams, 2006), which helped raise concern about acid rain, and (later) protecting the ozone layer and combating climate change. Concern within the EU’s energy policy domain for environmental action also began to grow during the 1980s, and EU policy on climate also dates from the 1980s. Attempts to link climate and energy policy thus dates from the emergence of climate change onto the political agenda in the late 1980s.

EU climate policy has rather developed alongside international agendas on climate negotiations since the 1980s. The Parliament was the first of the EU institutions to call explicitly for common policy measures to combat climate change (Jordan & Rayner, 2010; Wurzel & Connelly, 2011a). In 1986, the Parliament adopted its Resolution on “measures to counteract the rising concentration of carbon dioxide in the atmosphere (the ‘greenhouse’ effect)” (OJ C255/272 13.10.86). Furthermore, the agreement of the Single European Act (SEA) included a specific requirement that environmental requirements (including climate change) “shall be a component of the

42 Context and background

Community’s other policies” (Art. 130r). This article can be seen as calling for a level of environmental and climate policy integration in any energy policy developed (see also chapter three). It represents a legal commitment by the EU to advance EPI and later evolved into Article 11 of the TFEU.26

In 1988, action at the international level pushed climate change higher on the political agenda. In June 1988, an international conference took place in Toronto on “the changing atmosphere” (Pallemaerts & Williams, 2006, p. 23). The World Meteorological Organisation (WMO) and the United Nations Environment Programme (UNEP) agreed to set up the Intergovernmental Panel on Climate Change (IPCC) that same year, which aimed to engender consensus on climate science (Oberthür & Pallemaerts, 2010a, p. 29). In the EU, the Commission released its first communication on climate change suggesting policy options to respond to the “greenhouse effect” (European Commission, 1988). The options suggested included the recommendation that the EU and its member states take climate change “into account” in their policy decisions, especially related to energy (an early call for a certain degree of CPI); enhanced measures on energy savings and energy efficiency; and, sustaining research programmes on climate change (European Commission, 1988, p. 11). The European Council took up the environmental concern in its Rhodes declaration on the environment (also in 1988). In this declaration, the European Council mentioned the greenhouse effect and provided political backing for action to combat climate change (European Council, 1988; see Table 2.1).

2.2.2 Baby steps: EU climate and energy policy in the 1990s With the rise of the climate issue on the international and EU political agenda, and a general concern for environmental issues growing over the previous years, the Commission found the opportunity to develop further its competence on energy policy in making stronger links to environmental and internal market policy. Its 1990 Communication on “energy and the environment” was the “first time” that energy policy addressed “environmental problems” in an international context (European Commission, 1990, para. 1; interview 22). Especially important for energy policy in the 1990s was the Commission’s 1995 green paper on energy policy (European Commission, 1995). Here, the Commission outlined the “trinity” of EU energy policy

26 See Annex II for the full text of Article 11, TFEU.

43 Context and background

objectives (Pointvogl, 2009, p. 5708) as competitiveness, security of supply and environmental sustainability. These three objectives have remained at the core of EU energy policy since (see Table 2.2).

Period Energy Policy Climate Policy External events 1990s 1990: 1990: 1990: “Energy and the environment”, June: European Council calling First IPCC report COM (89) 369 for adoption of climate targets published 1993: S October: Joint Energy & 1992: Council Directive 93/76/EEC Environment Council meeting UN Rio Earth to limit carbon dioxide S calling for stabilisation of CO2 Summit adopts S emissions by improving energy emissions in EU by 2000 S UNFCCC efficiency (SAVE) 1992: Maastricht Council Decision 93/500/EEC Proposal for package of measures Treaty concerning the promotion of to combat climate change 1996:S S renewable energy sources in including: “Proposal for a Council Second IPCC report the Community (ALTENER) Directive introducing a tax on published 1995: carbon dioxide emissions and 1997: “An Energy policy for the energy”, COM (92) 226 Kyoto Protocol European Union”, COM (95) 682 1993: (KP) adopted 1996: Council Decision 93/389/EEC for a S Amsterdam Directive 96/92/EC concerning monitoring mechanism of Treaty common rules for the internal Community CO2 and other S S market in electricity greenhouse gas emissions Council Decision 96/391/EC 1997: laying down a series of “Climate change – the EU approach S measures aimed at creating a for Kyoto”, COM (97) 481 more favourable context for the 1998: development of trans-European EU agrees on burden sharing to networks in the energy sector meet 8 per cent GHG 1997: S emissions reduction target by “Energy for the future: 2008-2012 compared to 1990 renewable sources of energy”, levels S COM (97) 599 “Towards an EU post-Kyoto “The energy dimension of strategy”, COM (98) 353 climate change”, COM (97) S S 196 1998: Directive 98/30/EC concerning common rules for the internal market in natural gas Table 2.2: Some of the main policy developments in EU energy and climate policy in the 1990s (with agreed Directives, Decisions or Regulations in italics).

Sources: http://europa.eu; Birchfield & Duffield, 2011; Buchan, 2009; Damro, Hardie, & MacKenzie, 2008; Jordan & Rayner, 2010; Oberthür & Pallemaerts, 2010b; Oberthür & Roche Kelly, 2008; Pallemaerts & Williams, 2006; Wurzel & Connelly, 2011b. The Maastricht Treaty (1992) gave the EU competence on cross-border energy infrastructure, which was developed under the trans-European networks for energy (TEN-E) programme (see also chapter six; Buchan, 2010, p. 360). But the legislative focus in energy policy in the 1990s was on developing the internal energy market, beyond infrastructure requirements – a process that would continue for many years into the future. The so-called “first package” of legislative measures came with the

44 Context and background adoption of Directives 96/92/EC concerning common rules for the internal market in electricity and 98/30/EC concerning common rules for the internal market in natural gas, which both took several years to negotiate (Eikeland, 2011, p. 13). Persistent problems of implementation, and the dominance of large incumbents in the energy market led to this legislation being amended (twice) in the 2000s (see below; Pointvogl, 2009). Besides these policy developments, some limited energy policy developments in the 1990s came in response to the challenge of climate change.

The EU continued to develop its climate policy alongside the international climate negotiations agenda in this decade. The early 1990s involved policy preparation leading to the 1992 Rio Earth Summit, where the UNFCCC was opened for signature. EU internal preparations for this conference involved (some) member states developing their own GHG emission reduction plans, but efforts were made at the EU level to develop common policies and measures also (Jordan & Rayner, 2010; Oberthür & Pallemaerts, 2010a). In June 1990, the European Council called for the adoption of strategies and targets to limit GHG emissions, and for the Commission to develop proposals for concrete action to reduce emissions (European Council, 1990, Annex II). In October 1990, the first joint Council of energy and environment ministers took place, and ministers responded to the European Council’s call by adopting the target of stabilising CO2 emissions by 2000 at 1990 levels. The Commission followed with its first package of proposals to combat climate change, although this package was not adopted by the Council in time for the Rio conference in 1992 (Skjærseth, 1994, pp. 31-32).

This first package of policy measures to tackle climate change included four main legislative proposals (crossing into energy policy fields also):

• A framework Directive under the SAVE programme to conserve energy and improve energy efficiency (adopted in 1993 as Council Directive 93/76/EEC to limit carbon dioxide emissions by improving energy efficiency; see also chapter five); • A Decision to support renewable energy development (ALTENER projects);

• A Decision for a monitoring mechanism for CO2 emissions; and

• A Directive on a combined CO2/energy tax (Oberthür & Roche Kelly, 2008; Skjærseth, 1994; Wettestad, 2000).

45 Context and background

Of the four proposals, the CO2/energy tax was by far the most controversial, with division even within the Commission on the issue, and opposition from several member states. Taxation Commissioner Christiane Scrivener was opposed to the measure and clashed with environment Commissioner Carlo Ripa di Meana on the issue (Skjærseth, 1994, p. 28). Neither was Commission President Delors fully supportive of Ripa di Meana pushing for strong policy, based on considerations that this was contrary to the subsidiarity principle of policy being made at the lowest level of governance (Barnes, 2011, p. 46). Although the proposal for a CO2/energy tax survived discussions in the Commission, it then faced resistance in the Council, where unanimity was the required decision-making procedure. Several member states, and a strong business lobby (Skjærseth, 1994, p. 28), built opposition to the CO2/energy tax. The proposal was blocked by Council and eventually withdrawn by the Commission in 2001 (European Commission, 2001a). The EU’s climate policy in the mid-1990s thus amounted to little more than a monitoring mechanism and two programmes on energy efficiency and renewable energies that had been substantially weakened in the policymaking process (Oberthür & Pallemaerts, 2010a; Oberthür & Roche Kelly, 2008, p. 40).

In the run-up to international climate negotiations in Kyoto in 1997 there was again an internal push in the EU to demonstrate its leadership credentials on climate change by adopting targets and strategies, although this did not result in concrete policy developments. In June 1996, the environment Council declared the objective – in line with IPCC recommendations – that global temperature “should not exceed 2 degrees above pre-industrial level” (Council, 1996, climate change conclusions para. 6), an objective that has remained a guiding principle for climate policy since. The EU also proposed to reduce its emissions of CO2, methane (CH4), and nitrous oxide (N2O) by 15 per cent by 2010 compared to 1990 in advance of the Kyoto negotiations, which was a far more ambitious proposal than other parties at the negotiations were prepared to suggest (Oberthür & Pallemaerts, 2010a). In the end, the EU agreed to reduce its emissions of six GHGs – in addition to the three mentioned above, perfluorocarbons

(PFCs), hydrofluorocarbons (HFCs) and sulphur hexaflouride (SF6) – by eight per cent. Meeting this goal entailed agreeing a “burden sharing agreement” among EU- member states to achieve the eight per cent emissions reduction goal collectively. This agreement was reached in Council in June 1998 and adopted into law by

46 Context and background

Decision 2002/358/EC ratifying the Kyoto Protocol (Council, 1997; Haug & Jordan, 2010; Lacasta, Oberthür, Santos, & Barata, 2010). The burden sharing agreement foresaw GHG emission reductions of up to 28 per cent (Luxembourg) and emission increases of 27 per cent (Portugal), arriving at an overall reduction of eight per cent in the EU as a whole for 2008-2012, compared to 1990 levels (Lacasta, et al., 2010, p. 96).27

In all, the 1990s saw limited concrete movement forward on EU policies and measures in climate policy (see Table 2.2). Energy policies proposed under the climate package were significantly watered down in the policymaking process (e.g. SAVE and ALTENER). The adoption of the Kyoto Protocol in 1997, however, would require specific measures to meet the agreed eight per cent emissions reduction objective between 2008 and 2012 (the Kyoto Protocol commitment period), compared to 1990 levels. Such policy measures were finally agreed in the EU in the 2000s.

2.2.3 The 2000s: moving to credible common policies and measures Achieving competence at the EU level on energy policymaking (at least for the internal dimension) seemed finally to be an attainable goal in the 2000s. The Commission produced another green paper on energy in 2000 (following its 1995 energy green paper), this time titled “towards a European strategy for the security of energy supply” (European Commission, 2000d). The Commission warned of the risks behind the growing energy import dependence of the EU. The green paper re- emphasised the three-pronged objectives of energy policy in the EU and highlighted the potential co-benefits for security of supply and competitiveness of energy prices of taking action on climate change (i.e. by increasing energy efficiency and indigenous renewable energies; Buchan, 2009; European Commission, 2000d, parts one and two).

27 At this time the EU included 15 member states that were obliged to meet the Kyoto Protocol target.

47 Context and background

Period Energy Policy Climate Policy External events 2000s 2000: 2000: 2001: “Towards a European strategy for “Towards a European Climate US withdraws energy supply security”, COM Change Programme”, COM (2000) from KP & EU (2000) 769 88 S agrees to pursue 2001: 2003: ratification Directive 2001/77/EC on the Directive 2003/87/EC establishing Third IPCC promotion of electricity produced a scheme for greenhouse gas report published from renewable energy sources emission allowance trading within S Marrakech 2002: the Community accords Directive 2002/91/EC on the energy 2004: S Nice Treaty performance of buildings Directive 2004/101/EC on the 2005: 2003: linking of the EU ETS with S KP enters into Directive 2003/54/EC S project mechanisms under the force concerning common rules for Kyoto Protocol S Rejection of S the internal market in Decision No 280/2004/EC Constitutional electricity and repealing concerning a mechanism for S Treaty in French Directive 96/92/EC S monitoring Community and Dutch Directive 2003/55/EC greenhouse gas emissions and referenda concerning common rules for implementing the Kyoto 2006: S the internal market in natural Protocol Natural gas gas and repealing Directive 2006: crisis (especially 98/30/EC Directive 2006/40/EC on reducing S Eastern Europe) 2004: the emission of fluorinated Stern report Directive 2004/101/EC on the greenhouse gases published promotion of cogeneration 2007: 2007:S 2006: European Council agrees 20 per Fourth IPCC Directive 2006/32/EC on cent GHG emissions reduction by report published energy end use efficiency and 2020 target S Lisbon Treaty S energy services 2008: 2009: “A European strategy for January: climate and energy NaturalS gas crisis sustainable, competitive and package presented (especially Eastern S secure energy”, COM (2006b) S December: European Council, Europe) 105 Parliament and Council agree 2009: 2007: S on climate and energy package COP15 at “An energy policy for Europe”, 2009: Copenhagen, COM (2007) 1 Directive 2009/29/EC Denmark 2008: amending Directive 2010: COP16 in “Second strategic energy review. S 2003/87/EC so as to improve Cancun, Mexico An EU energy security and and extend the greenhouse gas solidarity action plan”, COM emission allowance trading (2008) 781 scheme of the Community 2009: Decision No 406/2009/EC on Directive 2009/28/EC on the the effort of member states to promotion of the use of energy from S reduce their greenhouse gas renewable sources emissions to meet the 2010: Community’s greenhouse gas Directive 2010/31/EU on the emission reduction energy performance of buildings commitments up to 2020 Directive 2009/31/EC on the geological storage of carbon S dioxide Table 2.3: Some of the main policy developments in EU energy and climate policy in the 2000s (with agreed Directives, Decisions or Regulations in italics).

48 Context and background

The EU adopted further legislation to make steps towards the completion of the internal energy market. This “second package” was adopted in 2003 as Directives 2003/54/EC (repealing Directive 96/92/EC on electricity) and 2003/55/EC (repealing Directive 98/30/EC on natural gas; Dupont & Primova, 2011, pp. 10-12; Eikeland, 2011, pp. 19-21). This step forward in the completion of the internal market opened up competition in electricity supply for all consumers by 2007. The “third package” followed in 2009. Directives 2009/72/EC (on electricity) and 2009/73/EC (on natural gas) required the separation of supply and generation companies from transmission companies (Eikeland, 2011, pp. 21-26). The negotiations leading to these pieces of legislation were difficult and resulted in three models of so-called “unbundling” being finally agreed – including the “ITO” (independent transmission operator) model that allowed vertically integrated companies “to keep their ownership of the transmission system provided they are managed by an independent transmission operator” (Dupont & Primova, 2011, p. 10). These policy measures represent the major pieces of legislation for advancing the single EU energy market and enhancing EU energy policy in the 2000s.

Along with the early energy and climate policy developments, including the 2001 Directive on the promotion of electricity produced from renewable energy sources (2001/77/EC; see chapter four) and the 2002 energy performance of buildings Directive (2002/31/EC; see chapter five), moves towards a common EU energy policy gained a boost after the informal European Council Hampton Court summit in 2005 (Buchan, 2009; Pointvogl, 2009; interview 22). This meeting was called by then UK Prime Minister Tony Blair during the UK Presidency, at a time when UK energy independence was diminishing and concern for climate change was increasing. Blair made energy policy a central element of the UK Presidency (Buchan, 2009, p. 8). It followed the EU’s major enlargement in 2004 that saw an increase from 15 to 25 member states at the negotiating table (with two more joining in 2007). Agreement was reached at this meeting “to move forward work in the energy sector and to establish a common European grid” (Blair, 2005).

In 2006, the EU suffered an energy security crisis when Russia and Ukraine engaged in a natural gas price dispute and gas supplies to Ukraine were cut. This dispute affected supplies to several EU countries – particularly the new member states in Central and Eastern Europe (Pirani, 2012). While energy security was always part of

49 Context and background the energy policymaking agenda, this event served to focus on the security of the EU’s energy supplies, as the EU is an import-dependent bloc (see also chapter six).

Following the Hampton Court summit and the gas crisis in early 2006, the Commission released yet another green paper on energy policy discussing “a European strategy for sustainable, competitive and secure energy” (European Commission, 2006b). This green paper was followed closely by documents that outlined the future integrated climate and energy policies: the communications “an energy policy for Europe” (European Commission, 2007c) and “limiting global climate change to 2 degrees Celsius: the way ahead for 2020 and beyond” in 2007 (European Commission, 2007e). EU climate policy development was now more closely linked to energy policy development.

In 2008, the Commission’s second strategic energy review touched again upon the importance of energy security (European Commission, 2008e), which highlighted a certain clairvoyance in the Commission in light of the ensuing gas crisis between Russia and Ukraine in early 2009 (Pirani, Stern, & Yafimava, 2009; see also chapter six). However, despite these many developments in energy policy and despite the new energy chapter in the Lisbon Treaty (Art. 194), which entered into force in late 2009, the EU’s competences on energy policy remain rather weak (Jordan & Rayner, 2010, p. 72). Article 194 of the TFEU lists the four areas where energy policymaking competence resides at the EU-level:

• Ensuring the functioning of the energy market; • Ensuring the security of energy supply; • Promoting energy efficiency and energy saving, and the development of renewable forms of energy; and • Promoting new network interconnections.

While this may seem like a substantial expansion of competence, there are two caveats that limit EU competence. These caveats are set out in paragraph two of the same Article: “such measures shall not affect a member state’s right to determine the conditions for exploiting its energy resources, its choice between different energy sources and the general structure of its energy supply”, and in paragraph three: “the Council…shall unanimously…establish the measures…when they are primarily of a

50 Context and background fiscal nature” (TFEU, Art. 194(2); 194(3)). The procedural limits that helped block the CO2/energy tax in the 1990s remain in place. The energy chapter seems to codify the policy developments that had already happened over the previous decades “organically” through interlinkages with other competences (environment or internal market), rather than to specify potential new areas for EU energy policymaking (Buchan, 2009, pp. 8-9).

The EU thus seems to have spurred action on energy policy in the 2000s based on previous developments. In the same decade, however, there was a great leap forward in climate policymaking. Climate policy internally in the EU in the 1990s was slow in being negotiated and often watered down substantially during the negotiations (or abandoned, as with the CO2/energy tax). But a renewed interest in internal climate policy development in the EU, especially in the mid-2000s, meant that often far- reaching policy could be proposed. This can in part be understood as the EU’s attempts to make good on its claims for international leadership on climate change (Boasson & Wettestad, 2013; Jordan & Rayner, 2010; Oberthür & Pallemaerts, 2010a; Oberthür & Roche Kelly, 2008), but also due to a growing citizen concern for climate change (Eurobarometer, 2008, 2011).

At the turn of the century, there was a sense of frustration in the Commission at the poor record and slow and difficult internal negotiations on climate (and energy) policy (Jordan & Rayner, 2010, p. 66). Discussions and debates among the EU institutions and among member states in the late 1990s delayed Commission proposals on increasing the shares of electricity from renewable energy sources, for example (Boasson & Wettestad, 2013, pp. 83-87; Howes, 2010; see also chapter four). In 2000, then, the Commission tried a new approach with the proposal for a “European Climate Change Programme” (ECCP) that would involve many actors and stakeholders in developing measures and policies to combat climate change (European Commission, 2000a). During the first ECCP (from 2000 to 2004), eleven working groups, drawing on experts from the Commission, member states, industry and NGOs, aimed to identify options for reducing GHG emissions. These working groups covered:

• Emissions trading;

51 Context and background

• Joint implementation and clean development mechanism (Kyoto Protocol flexible mechanisms for meeting GHG emissions reductions targets); • Energy supply; • Energy demand; • Energy efficiency in end-use equipment and industrial processes; • Transport; • Industry; • Research; • Agriculture; • Sinks in agricultural soils; • Forest-related sinks.28

On the international stage, a major upheaval in climate policy came in 2001 with the announcement from the George W. Bush administration that the US would not ratify the Kyoto Protocol, calling the Protocol “unfair” and “ineffective” (letter from President George W. Bush to Senators Hagel, Helms, Craig and Roberts, 13 March 200129). This led to much discussion in the EU about whether or not to move forward with ratification itself. Two criteria needed to be fulfilled before the Kyoto Protocol could enter into force: first, it needed to be ratified by at least 55 parties to the UNFCCC, and those parties needed to represent 55 per cent of industrialised countries’ CO2 emissions in 1990 (Art. 25(1), Kyoto Protocol). Without ratification by the United States, meeting these two criteria became more difficult. But the EU decided to work to save the Kyoto Protocol (Oberthür & Roche Kelly, 2008, p. 36). The EU was instrumental in ensuring the adoption of the Marrakech Accords on the implementation of the Protocol in 2001, and it ratified the Kyoto Protocol in 2002 (Damro, et al., 2008, p. 180). Diplomatic efforts continued over the next number of years to ensure enough parties ratified the Protocol to ensure its entry into force. Once Russia ratified in late 2004 – after the EU promised to back its request for WTO membership – the Protocol finally entered into force in 2005 (Pallemaerts & Williams, 2006, p. 41).

28 See: http://ec.europa.eu/clima/policies/eccp/first/index_en.htm, accessed: 17 February 2013.

29 Available at: http://georgewbush-whitehouse.archives.gov/news/releases/2001/03/20010314.html, accessed: 19 February 2013.

52 Context and background

During these early years of the twenty-first century, the EU advanced internally on climate policy also. The centrepiece of EU climate policy was the 2003 Directive on the development of the emissions trading system (ETS; Directive 2003/87/EC). This was a cap-and-trade scheme for CO2 emissions from major industrial and energy installations in the EU. The cap (or ceiling) for emissions under the system was linked to the EU’s commitment to reduce GHG emissions under the Kyoto Protocol (van Asselt, 2010, p. 128). The first (pilot) phase of the ETS ran from 2005-2007, and already demonstrated weaknesses in the system, due predominantly to an over- allocation of free emission allowances to installations. Later reformulations of the ETS (see below) aimed to reduce the cap and also the amount of free allowances provided to installations (Boasson & Wettestad, 2013; Skjærseth & Wettestad, 2008; Wettestad, 2005). The agreement on the 2003 ETS Directive was followed with the 2004 “linking Directive” (2004/101/EC) that linked the ETS to the flexible mechanisms of the Kyoto Protocol – namely the clean development mechanism (CDM), which allows industrialised countries to gain credits by investing in GHG emission reduction projects in developing countries, and joint implementation (JI), which is a similar mechanism for projects in economies in transition (Oberthür & Pallemaerts, 2010a). Other policy measures during these early years of the decade to help reduce GHG emissions were taken in energy policy, such as the Directives on renewable sources of electricity, energy performance of buildings, among others (see Table 2.3, chapters 4 and 5).

The EU entered a period of heightened political commitment to taking action on climate change from about the time of the entry into force of the Kyoto Protocol (Boasson & Wettestad, 2013, p. 43). The Commission prepared for proposing far- reaching and ambitious policies that would advance the international negotiations on the post-Kyoto Protocol framework. A major political advancement for internal EU climate policy came with the European Council’s endorsement of a goal to reduce GHG emissions in the EU by 20 per cent by 2020 in March 2007. While the European Council called on developed countries collectively to agree to reduce emissions “in the order of 30% by 2020 compared to 1990”, the EU made a “firm independent commitment to achieve at least a 20% reduction of greenhouse gas emissions by 2020 compared to 1990”. In addition, the European Council agreed that the EU endorsed the 30 per cent target for itself by 2020 “provided that other developed countries

53 Context and background commit themselves to comparable emission reductions and economically more advanced developing countries to contributing adequately according to their responsibilities and respective capabilities” (European Council, 2007, p. 12, paras. 30- 32). This signalled a concrete political commitment from EU leaders to advance, not only internal EU climate policymaking, but also international climate negotiations.

The Commission’s response to the commitment of the European Council was to propose a package of policy measures – the “integrated” climate and energy package. The package of measures, proposed in January 2008 (European Commission, 2008a), included proposals on renewable energy, carbon capture and storage (CCS), a proposal for an effort-sharing decision among member states to reduce GHG emissions not covered by the ETS, and revisions to the ETS Directive. Negotiations were rapid (Boasson & Wettestad, 2013, p. 48) and agreement on the package was reached in the European Council in December 2008 (European Council, 2008b, p. 8). Council and Parliament also agreed to the measures in first reading in December 2008 and they were adopted into EU law in 2009 as Directives 2009/28/EC on renewable energy (see chapter four), 2009/31/EC on CCS, and 2009/29/EC on the ETS. A new Decision on “effort sharing” among member states (evolving from the “burden sharing” 1998 agreement) on the reduction of emissions from sectors not covered by the ETS was adopted as Decision 406/2009/EC. In addition, Regulation (EC) No

443/2009 on mitigating CO2 emissions from cars, the third internal energy package (see above) and Directive 2010/31/EU on the energy performance of buildings (see chapter five) were also negotiated internally during this time (see Table 2.3).

The climate and energy package marked a period of high political commitment to action on climate change, and to promoting binding legislative measures (with the notable exception of the non-binding 20 per cent energy savings target by 2020). From 2005, with renewed emphasis on energy policy, the entry into force of the Kyoto Protocol, and the rejection of the EU’s constitutional treaty in referenda in France (in May 2005) and in the Netherlands (in June 2005), a “window of opportunity” to move forward on climate and energy policy was opened (Kingdon, 2003, p. 166) – climate change was seen to provide an opportunity to advance EU integration generally, and to project EU “leadership” on the issue onto the international stage (Oberthür & Roche Kelly, 2008, p. 43). With the financial and economic crises that have gripped the EU in the years since 2008, this window may

54 Context and background have begun to close, but not before the climate and energy package was adopted in 2009.

2.3 Conclusions This chapter set out the background and context for a study of CPI in the EU’s energy policy. First, I discussed the policy process and the decision-making actors in the EU (and their role in climate and energy policymaking). Second, I outlined the main developments in climate and energy policy over time.

The policymaking process in the EU is a rather complex procedure involving several actors and institutions. The main legislative procedure is the co-decision procedure (renamed the “ordinary legislative procedure” in the TFEU), which involves the Commission, the Parliament and the Council as the three co-deciding institutions. For the purposes of CPI, it is also relevant to look to the European Council as an important institution, as well as other stakeholders that may attempt to influence the policymaking process. In energy and climate policy, the EU’s competence to develop policy has been mixed and linked more to competence on the internal market and on the environment than on energy. This has changed slowly over time, along with the reform of the Treaties, but also as a codification of the practice of policymaking that developed organically.

What the historical overview of EU energy and climate policy has shown is that, although energy policy has been part of the European project from its origins, it has developed rather slowly, with member states long holding onto power and sovereignty on energy decisions in the EU. Climate policy, however, made comparatively swift progress from the moment it came onto the political agenda in the late 1980s. Although there were occasions where policy development was slow (or non-existent) at the EU-level (especially in the early 1990s), climate policy in the EU leaped forward in the mid-2000s. Much of energy policy during these times developed under the EU’s competence in the internal market, but also under its environmental competence.

The history of EU climate and energy policy is therefore linked. From the emergence of the climate issue on the political agenda, it was linked to energy policy measures. The link between the two sectors seems to have been concretised in the climate and

55 Context and background energy package adopted in 2009. The main policy measures in place to combat climate change to 2020 include a Directive on increasing the share of renewable energy in the EU. Several other policy measures under the energy sector have climate benefits (such as measures on improving energy efficiency). Such developments seem to point to certain levels of CPI in EU energy policy. However, this historical overview is not enough to say that there are high levels of CPI in the EU’s energy sector. In-depth analysis and a clear conception of how to operationalise and explain CPI is required.

Conceptual framework

Answering the two-step research question outlined in chapter one (“what is the extent of climate policy integration in EU energy policy, and how can this level of CPI be explained?”) requires a clear conceptual and theoretical framework. This involves conceptualising and operationalising CPI to be able to measure its extent in the policy process and output, and also developing a framework to explain the results of the study. This chapter outlines the theoretical framework guiding the empirical study and analysis described in chapters four to six.

In this chapter, I first review the literature on policy coherence, coordination and integration, before discussing the theories and explanatory variables used in environmental policy integration (EPI) literature (the main body of literature for a study on CPI). This results in an identification of how CPI is conceptualised and operationalised (section 3.1). A study of CPI in the EU context can make use of general theories of European integration, which can explain the EU policy process. In section 3.2, I discuss briefly three theories from which I derive a set of explanatory variables (as informed by the literature on EPI). The four main explanatory variables identified for this study are: functional interrelations; political commitment; institutional and policy context, which each explain CPI in both the policy process and output; and the process dimension, which explains CPI in the policy output only.

3.1 Conceptualising climate policy integration To conceptualise CPI, it is necessary to situate the concept within the academic literature. Concepts such as policy coherence and policy coordination are linked to CPI. However, as climate policy has traditionally been part of the environmental policy field, CPI is firmly rooted in the literature on EPI. I thus necessarily devote more space to a review of the vast body of literature on EPI.

3.1.1 From policy coherence and coordination to policy integration Policy coherence and policy coordination have long been promoted as tools for enhancing effective and efficient policymaking as part of wider policy analysis

57 Conceptual framework literature (Mickwitz et al., 2009; Peters, 1998; Sabatier, 2007). Policy integration takes both these literatures into account and is a concept that engenders a holistic view of improving efficiency in policymaking (and effectiveness in policy output and implementation). The literature around the concept of policy integration also has a relatively long history (Underdal, 1980, being a seminal early text on policy integration).

As key concepts within the policy integration discussion, “policy coherence” and “policy coordination” focus on and emphasise slightly different aspects of integration. Policy coherence aims to avoid conflicts between the objectives of different policy areas (and to promote as far as possible “win-win” solutions; van Bommel & Kuindersma, 2008, p. 15). In other words, by promoting policy coherence, policy sectors would interact to agree on effective and efficient policies that produce successful combined results. Policy coherence thus emphasises the final policy output – i.e. that policy outputs be coherent with each other, and not contradictory or counterproductive (OECD, 2002).

In general, policy coordination can be seen to aim for the same results, namely that “policies and programmes of government are characterised by minimal redundancy, incoherence and lacunae” (Peters, 1998, p. 296), although the emphasis is rather on achieving this through coordination efforts during the policy process. Policy coordination has been described as an “administrative Holy Grail” (ibid., p. 295), implying that achieving policy coordination is a long-held challenge for policymakers. Metcalfe (1994) suggests a scale of coordination that represents both a set of tasks for the achievement of policy coordination, and also a tool for identifying the level of policy coordination in any given polity. The scale ranges from low levels of policy coordination, where independent decision-making in different policy sectors takes place, to high levels of policy coordination, where an overall government strategy is established (ibid., p. 281). The full scale of policy coordination levels (and tasks) is outlined in Figure 3.1. Policy coordination clearly deals predominantly with improving the policy process and ensuring coordination takes place during the negotiation and policy development in and among policy sectors.

58 Conceptual framework

Figure 3.1: Increasing scale of policy coordination.

Source: Metcalfe, 1994, p. 281.

Policy integration encompasses both coherence and coordination by emphasising both the policy process and the policy output. Some analysts and researchers may contend that policy coordination and coherence are synonymous (Mickwitz, et al., 2009, p. 24), but from further investigation, a clearer understanding is that policy coherence and coordination refer to different aspects of the policy cycle: while promoting policy coherence implies ensuring various policy outputs are harmonious, achieving policy coordination implies putting in place communication and coordination mechanisms to improve efficiency (leading to improved effectiveness) in the policy process. Zingerli, Bisang and Zimmermann (2004, p. 3) also express the view that policy coordination is just one part of the overarching concept of policy integration – and can be regarded therefore as a “degree of integration”. Accordingly, policy integration can be viewed as a combination of coherence and coordination to encompass the policy process and output (and also can be understood as a matter of degree itself, see below). The extent of policy integration overall may thus be established by examining both the policy process for evidence of the integration of the objectives of the second policy sector (in the case of this research project, the integration of climate policy objectives), and the policy output for synergy among policy objectives (i.e. climate policy objectives and the sector policy objectives interacting harmoniously or synergistically).

59 Conceptual framework

With a clear understanding of the importance of coherence and coordination within the overarching concept of integration, we can now turn to Arild Underdal’s rationalist conceptualisation of policy integration. Examining policy integration from the perspective of the policy cycle, he argues that an integrated policy “must meet three basic requirements, viz comprehensiveness, aggregation and consistency”, corresponding respectively to the input stage (policy preparation), the processing of inputs (policy negotiations) and the output stage of the policy cycle (Underdal, 1980, p. 159). In the terminology used in this research project, the comprehensiveness in the input stage matches the phase of the policy process where the European Commission assesses and proposes policy options (policy process); aggregation in the processing of inputs refers to integration during the inter-institutional negotiations and amendments to the proposals of the Commission (policy process); and, consistency in the output stage refers to the consistency between the final policy decision (policy output) and the objectives of climate policy.

There is no clear agreement in the literature about the ideal end result of policy integration. For example, integration of two or more policies could eventually lead to the creation of one unified overall policy or, alternatively, certain policy concerns and objectives could be integrated into one or more existing policies. This lack of agreement is clear when Underdal is compared to Briassoulis. Underdal explains that an integrated policy means “a policy where constituent elements are brought together and made subjects into a single, unifying conception” (ibid.), clearly promoting the idea that policies are integrated to become a new unified policy. However, Briassoulis disputes this single interpretation of the end goal of integration, highlighting in addition the aim of incorporating concerns into “an extant policy to produce an integrated policy” (2005a, p. 50).

In related terms, policy integration may be conceived of as a unidirectional or mutual process of adaptation. Underlying these differences are frequently different conceptions of the weighting and prioritisation of different policy objectives. Considering different policy objectives as equally valid may lead to pursuing mutual adaptation, the construction of an “aggregate measure of utility” (Underdal, 1980, p. 162) and the creation of an “integrated and coherent policy system” (Briassoulis, 2005a, p. 50). Giving priority to one particular policy objective may lead us to emphasise the unidirectional adaptation of individual existing policies to integrate this

60 Conceptual framework policy objective. These differences in approach also inform the debate about, and the conceptualisation of, environmental and climate policy integration, which is discussed next.

3.1.2 From environmental (EPI) to climate policy integration (CPI) Combating climate change has traditionally fallen under the remit of environmental policymaking. Although this has been changing,30 the history of climate policy development is rooted in environmental concerns. Thus, the literature on environmental policy integration is the natural home for the elaboration of CPI.

Research on EPI has quite a long history in the EU. With the first development of EU environmental policy in the 1970s, the concept began to be discussed in EU policymaking. The Single European Act (SEA, 1987) formalised the concept with the introduction of the objective that “environmental protection requirements shall be a component of the Community’s other policies” (Article 130r(2)). EPI became an even more developed concept after the Amsterdam Treaty (1997) outlined sustainable development as an overall objective of the EU. In addition, Article 6 of the Amsterdam Treaty (now Article 11 of the Treaty on the Functioning of the European Union, TFEU) even commits the EU by stating, “environmental protection requirements must be integrated into the definition and implementation of the Union policies and activities, in particular with a view to promoting sustainable development”. Policy integration is thus a developed and familiar topic in the EU. It is a legal objective of the EU, yet its practical application in policymaking is varied (Bryner, 2012; Jordan & Lenschow, 2010).

The EU may well be the jurisdiction that has demonstrated most willingness and dedication to advance EPI. In its first Environmental Action Programme (EAP),31 although the Commission did not explicitly mention EPI, it advanced environmental principles such as the precautionary principle and the “polluter pays” principle (European Commission, 1973; Lafferty & Hovden, 2003, p. 3). The Commission also discussed the notion of evaluating other policies for environmental effects in this EAP,

30 Some government administrations now have ministries or cabinets on climate change separated from environment, such as the UK with its department on energy and climate change, and the European Commission with its Directorate-General (DG) on Climate Action in place since February 2010.

31 The first EAP was adopted in the EU in 1972. The period 2012-2020 is covered by the EU’s seventh EAP.

61 Conceptual framework and was the first EU institution to push for EPI (Lafferty & Hovden, 2003, p. 3; Lenschow, 2002a, p. 3). However, even with this early commitment from the Commission, and the later codifications of the EPI principle into EU law under the SEA, Maastricht and Amsterdam Treaties, concrete responses to implementing EPI in policy were rather disappointing in the EU (Jordan, Schout, & Unfried, 2008, pp. 162- 164; Wilkinson, 1998, 2007). Early attempts to advance EPI included soft coordinating measures, yet throughout the 1980s and 1990s, very little advancement was made. The 1998 “Cardiff Process”, which aimed to ensure that EU institutions (and especially the Council and the European Council) engaged in common reporting and policy reviewing for EPI, fizzled out without having lived up to expectations of advancing EPI (Jordan, et al., 2008, pp. 163-164). Neither have institutions nor actors beyond DG Environment in the Commission actively engaged with the objective of EPI in the EU (ibid., p. 165). This may be partly explained by the conceptual vagueness of EPI, and the difficulties in defining it. The literature on this topic is quite disparate – with different definitions, perspectives and explanatory variables. Indeed, EPI as a concept has thus far escaped definitional consensus and clarification.

EPI emerged and developed as a concept because of the perceived failure and inability of traditional sectoral environmental policymakers to respond to the pressures placed on the environment by the actions of a wider society. It can be considered a broad concept that is value-laden and political, thus resulting in several interpretations (Persson, 2007, p. 44). For these reasons, there is much debate on the meaning of the term, how it can be defined, recognised, implemented, measured, evaluated and improved upon. In fact, the European Environment Agency describes how the lack of agreement on the meaning of EPI has helped to make it “more acceptable to policy-makers” (EEA, 2005, p. 12).

What is clear, however, is that EPI has a normative dimension in favour of the environment. Placing an adjective before the term “policy integration” implies assigning preference, weight or priority to a specific policy sector’s objectives over another (Briassoulis, 2005b, p. 23). As outlined above, it is clear that European law reflects this normative dimension, supporting in particular the integration of environmental objectives into other policy sectors.

62 Conceptual framework

What remains unclear is how much weight or priority the environment does or should receive in the policy process and its output. Reflecting the broader discussions about policy integration in general (see above), authors differ on whether they regard EPI as a policy principle; a process or a set of measures or institutional procedures; a policy output or outcome; or some combination of these (Persson, 2004, pp. 22-25). In this context, they vary as regards the strength and depth of EPI they employ as a standard and benchmark. Some have advocated that environmental objectives should receive “principled priority” in other policy sectors, throughout the policy process and, especially, in the final output (also called “strong” or ideal EPI; Lafferty & Hovden, 2003, p. 9). On the other end of the spectrum, others have emphasised the importance of simply taking environmental considerations into account in the formulation of policy in other sectors (“weak” EPI; Jordan & Lenschow, 2008a; Persson, 2004).

In conceptualising CPI, I take inspiration from the “strong” standard set by Lafferty and Hovden in the policy process and output. Their two-tiered definition of EPI is as follows:

• “the incorporation of environmental objectives into all stages of policymaking in non-environmental policy sectors, with a specific recognition of this goal as a guiding principle for the planning and execution of policy; • “accompanied by an attempt to aggregate presumed environmental consequences into an overall evaluation of policy, and a commitment to minimise contradictions between environmental and sectoral policies by giving principled priority to the former over the latter” (Lafferty & Hovden, 2003, p. 9).

Bryner provides a reasoned (and vague) definition of EPI as “a greater awareness of how policies in one area affect another [that] encourages policy makers to prioritise environmental concerns” (Bryner, 2012, p. 22). Even this looser definition provides a normative understanding in favour of the environment. Nevertheless, there are good reasons for accepting the unavoidably normative standard for EPI (and CPI) of “principled priority” as an ideal/strong level. Taking a stance is inescapable due to the inherently normative nature of the concepts of EPI and CPI as a standard of policy evaluation. Most importantly, the standard for operationalising CPI needs to be explicated to make it transparent, to facilitate comparison with other research results,

63 Conceptual framework and to enable criticism and a full understanding. Choosing a high standard (such as “principled priority”) for CPI has the advantage of being comprehensive in two dimensions. Firstly, it covers the policy process and its output, and secondly, it can arguably cover the full spectrum from strong to weak EPI. As we will see below, in operationalising CPI, a principled priority standard provides a benchmark for evaluation, which can lead to an understanding of the gap between the status quo and the “ideal” or strong standard of CPI.

On the basis of the above discussion of EPI, I argue that CPI is thus a matter of degree (Bryner, 2012). CPI can exist to lower or higher levels in the policy process and policy output of sector policies and an understanding of the concept of CPI that allows for the principled priority of climate policy objectives over the other policy sector objectives is rather an expression of a high, complete or ideal benchmark against which actual levels can be measured. Therefore, rather than defining CPI as assigning principled priority in policymaking, I suggest that the principled priority of climate policy objectives in all stages of the policy process and its output in non- environmental policy sectors is the highest level of CPI that can be achieved.

Additionally, principled priority as a definitional concept for CPI is problematic for sustainable development objectives. While taking Lafferty and Hovden’s definition of EPI as outlined above, and simply replacing the word “environment” with “climate” results in a reasonable definition (Ahmed, 2009, p. 11), this is problematic and counterproductive if the aim is to prevent irreversible damage to environmental life- support systems.32 Consequently, recent literature has tried to overcome the problem of assigning principled priority to one environmental objective over another by distinguishing between what has come to be called “external” EPI in non- environmental policy sectors, and “internal” EPI between environmental sub- policies.33 In this way, principled priority can be granted to environmental/climate objectives for cases of external EPI, while a balancing and synergy logic is applied to cases of internal EPI (Biermann, Davies, & van der Grip, 2009; Kulovesi, Morgera, & Muñoz, 2010; Oberthür, 2009). As I focus here on “external” CPI into the EU’s

32 For example, a policy that aims to combat climate change by promoting the plantation of fast growing trees may detrimentally affect biodiversity and habitat protection measures (see Dupont, 2010).

33 In the case of the integration of climate and environmental policies, the highest level of CPI/EPI that would be in line with sustainable development would aim for win-win outputs and synergy, as opposed to principled priority for climate objectives.

64 Conceptual framework energy sector, the strong standard of “principled priority” is appropriately applied as a benchmark of policy assessment.

3.1.3 Operationalising CPI The empirical application of the strong CPI standard requires an understanding of how to measure different levels of CPI, in both the policy process and its output. For the purposes of such a measurement, an overall guiding question can be asked in each case: what would the ideal/complete level of CPI in this sector look like? This question is examined for both the policy process and the policy output.

3.1.3.1 Measuring CPI in the policy process For both the policy process and output, I measure the level of CPI on a fivefold scale, from very low to very high over low, medium and high (see Tables 3.1 and 3.2 below). In measuring CPI in the policy process, I use a number of factors and sub-indicators to arrive at the final result.34

As outlined in chapter two, with the explanation of the co-decision procedure, CPI in the EU policy process involves all three deciding institutions (with political backing from the European Council) to ensure climate policy objectives are integrated into the sectoral policy under negotiation. Therefore, it is possible that very high levels of CPI would be found where there is ease of access of pro-climate policy stakeholders to the policy process. The involvement of pro-climate policy stakeholders in the policy process is an indicator of the attention paid to the climate policy issue during policymaking. In much EPI literature, emphasis is laid on the openness of the policy process to several voices to ensure environmental/climate issues are heard and dealt with (EEA, 2005; OECD, 2002; Jordan & Lenschow, 2008b). Hence, the involvement of pro-climate stakeholders in the policy process is considered here an indicator of the level of CPI in the policy process. This is broken down into three distinct operational indicators.

First, I assess the role played by internal climate policy stakeholders in the policy negotiations. This refers to the roles played by DG Environment (and later, DG Climate Action, if applicable) in the Commission, by the ENVI committee in the

34 The level of CPI in the policy process can also act as an explanatory variable when explaining the level of CPI in the policy output, as the level of CPI in the policy process can be expected to affect the level of CPI in the policy output. This is discussed in further detail in section 3.2.3 below.

65 Conceptual framework

Parliament, and by the Environment Council throughout the policy negotiations. To assess this, I look at the responsible DGs, committees and Council formations on each policy file, and if these are not the pro-climate policymakers, then assess whether or not these pro-climate actors were involved in the negotiations (and to what extent). These roles vary loosely on a qualitative five-fold scale, matching the scale for the level of CPI in the policy output, from “very high/complete” (when internal pro- climate stakeholders are in the lead on the policy proposal and negotiations); “high” (pro-climate stakeholders are co-drafting and/or co-deciding with other sectoral DGs, committees and Council formations); “medium” (they are consulted throughout the policy process); “low” (they are not consulted, but can or do provide unsolicited opinions); and “very low/none” (there is no involvement and no visibility of pro- climate stakeholders in the process). The five-fold scale is thus based on the conceivable possibilities for pro-climate stakeholder involvement on a certain policy file within the institutions. The different levels of “involvement”, in this case, are dictated by the procedural rules that could be applied in the EU’s co-decision procedure. Through an analysis of documents associated with the policy negotiations, supplemented with interview data, media reports and literature review, I can assign a score on the “involvement of internal pro-climate policy stakeholders” within the EU co-deciding institutions in the policy process.

Second, I look to the role played by external pro-climate policy stakeholders in the policy process – such as environmental NGOs and industrial actors in favour of stringent climate policy. The involvement of external pro-climate actors in the policy process can lead to heightened pressure on policymakers to ensure climate policy objectives are integrated into the development of policy (EEA, 2005; Lenschow, 2002b; Persson, 2007). I assess the access these stakeholders have to the policy process (and flowing therefrom, whether or not they may be able to influence the policy process). Access to the co-deciding institutions depends on two factors – first, the procedures for consultation that are in place in a certain institution, and second, the “openness” of the institution to input from external stakeholders. These two factors provide the basis for the fivefold scale for the measurement of the involvement of external pro-climate stakeholders in the policy process. This is assessed through an analysis of the official public consultations organised by the EU institutions, but also backed-up qualitatively with interviews with involved stakeholders who assess their

66 Conceptual framework own involvement in the policymaking process, with media reports and with a literature review. Such an assessment allows me to establish whether CPI in the policy process for this indicator can be “very high” (pro-climate stakeholders are invited by the EU institutions to provide input in the policy process; access is guaranteed and easy; and pro-climate stakeholders’ arguments receive wide backing); “high” (easy access of external pro-climate stakeholders to the process; strong pro- climate lobby that faces little opposition from other stakeholders or policymakers); “medium” (procedures allowing access, but the ease of access may be limited; the pro-climate lobby faces certain opposition from other stakeholders and from some policymakers); “low” (no or little access to the process; the pro-climate lobby opposed by majority of stakeholders and/or policymakers); and “none/very low” (no pro-climate stakeholders involved or visible in the policymaking process).35

Some recent literature discusses the general role of external stakeholders in EU energy policy. This can also inform a study on the level of CPI in the energy policy of the Union. EU energy policymaking has been found to be compartmentalised according to the topic, and skewed in favour of the participation of stakeholders from industry. In a study carried out of the main stakeholder consultation forums with DG Energy, Vasileiadou & Tuinstra found limited numbers of pro-climate actors, with energy companies especially, rather than environmental activists for example, playing a greater role in these forums (2012).36 As each forum has its own purpose and topic of discussion, the extent to which climate change is considered varies, with the issue of climate change often being bracketed under one forum dealing with renewables and energy efficiency measures (namely, the Amsterdam forum from 2005-2008, later becoming the Bucharest forum; ibid., p. 16). This may be understandable, due to the synergetic links between climate policy objectives and those of renewable energy and energy efficiency policy, but it does mean that climate objectives are not necessarily discussed where infrastructure or security of supply, for example, are the main topics

35 An argument can also be made that even without environmental or climate policymakers or stakeholders involved directly in the policy process, CPI can be advanced due to other actors being already convinced of the need for action on climate change. This is taken into consideration in the indicator that examines the recognition of sectoral functional interrelations as part of the analysis of CPI in the policy process. However, the strength of CPI in the policy process is, nonetheless, expected to be affected by the role of explicitly pro-climate voices.

36 The main consultation forums on energy include the Florence forum on electricity; the Madrid forum on gas; the Berlin forum on fossil fuels; the Amsterdam/Bucharest forum on renewables and energy efficiency; the Nuclear forum; and, the London Citizens forum.

67 Conceptual framework of focus. The authors of this study argue that such compartmentalisation of issues allows little room for cross-issue CPI, based on their assessment of the diversity of stakeholders, on the flexibility for change of a particular stakeholder forum, the frequency at which climate change is mentioned in meeting agendas and minutes, and an assessment of any measurable impact of stakeholder consultations on the policy process (ibid., pp. 6-8). Certain scholars have found that the compartmentalisation among stakeholder consultations is also reflected in the organisation of policymaking in the EU (Jordan & Lenschow, 2010; Jordan, Schout & Unfried, 2008, p. 161). Thus, this research emphasises the importance of pro-climate stakeholder voices for advancing CPI even in energy policy sectors that are not measures that can respond directly to the threat of climate change.

The issue of access for external stakeholders to the policy process deserves some further discussion. Measuring influence or impact on a policy process is notoriously difficult, methodologically (Adelle, Russel, & Pallemaerts, 2012; Hauser, 2011). Access to the policy process does not provide hard evidence that stakeholders have actually influenced the policy process and the resulting policy output, yet it is certain that influence is less likely to occur if the stakeholder has no access (Hauser, 2011, p. 62). Therefore, I will briefly discuss the ways and procedures under the co-decision procedure that allow access for external stakeholders to the policy process.

The Commission, as the initiator of legislative proposals, is the most-targeted institution for lobbying activities. Once a proposal is written, it can be difficult (but certainly not impossible) to amend it radically through the co-deciding procedures in the Parliament and Council (Tanasescu, 2009, p. 56). Therefore, many lobby interests aim to include their positions already in the policy proposal and seek to influence the Commission in its proposal writing. Access is often granted to the Commission, as it is understaffed and those staff may lack the technical expertise required for proposing legislation on specific issues (Hauser, 2011), but informal access to the Commission is often based more on personal relationships. Several pro-climate stakeholders expressed difficulties in accessing the right person at the Commission at the right time (interviews 1, 5, 6 & 7).

One method for external stakeholders to exert influence on the policy proposal is through the Commission’s use of expert groups. Expert groups are defined as

68 Conceptual framework

“consultative bodies that advise the Commission on the preparation of legislative proposals and policy initiatives, the implementation of legislation, programmes and existing Union policies, and the preparation of delegated acts”.37 Such expert groups generally consist of national and/or private-sector experts, but experts that are not member state representatives are more likely to come from business and industry, with little room for NGOs (Tanasescu, 2009, p. 56).38 Expert groups relevant for energy and climate policy include the Gas Coordination Group (with members coming from member states, energy and electricity associations, and industrial energy consumer groups, yet no environmental organisations), and the European Climate Policy Group (with members from the member states, from European Environment Agency and from neighbouring countries). While some authors agree that such expert groups may be practically useful for EU policymaking, they have also been criticised for their lack of transparency (and, thus, the difficulties in keeping such expert groups accountable) and the weight some of the groups may hold over policymakers (Broscheid & Coen, 2007; Gornitzka & Sverdrup, 2008). The general set-up and membership of these groups also does not privilege environmental and climate NGOs, although climate voices may already be present with some industrial members. The European Transparency Initiative (see below) launched in 2008 attempted to address some of the issues of transparency around lobbying in the EU (European Commission, 2009b).

The Commission, since 2002, also invites submissions through its online questionnaires that are available for all interested parties (citizens, business groups, NGOs, etc.).39 The responses to such public consultations ought to be considered in the policy proposal development and reported on by the Commission. Additionally, ad-hoc meetings organised by the Commission can be arranged to invite input from a range of stakeholders. These can be organised with a minimum of twenty days’ notice (Tanasescu, 2009, p. 85). Finally, the Commission’s occasional green papers can stimulate debate and consultation, not only with external stakeholders, but also within

37 This definition is taken from DG Internal Market and Services website: http://ec.europa.eu/dgs/internal_market/expert- groups/index_en.htm, accessed: 10 January 2013.

38 Member states may choose experts outside their ministries to represent their interests in expert groups if they wish (including academics, for example; interview 23).

39 Public consultation became a common procedure for the Commission after the adoption of its 2002 communication on consultation and dialogue (European Commission, 2002d).

69 Conceptual framework the EU institutions. However, there is no fixed pattern of consultation practices within the Commission, and DGs differ in how and to what extent they allow or invite input from external stakeholders (Gornitzka & Sverdrup, 2008; Tanasescu, 2009).

Commission consultation practices have, however, also developed over time and over the course of the decade under examination in this project: 2000-2010. With the white paper on European governance, published in 2001 (European Commission, 2001b), general principles for improving the EU policymaking process included more transparency, openness and the establishment of minimum standards for consultation on EU policy (ibid., p. 4). Following this paper, the Commission proposed the minimum consultation standards in 2002 (European Commission, 2002b), aiming to standardise practices of consultation throughout the different DGs of the Commission, and to ensure such practices were open and transparent. Thus, in later years, consultation procedures became more open and reported upon than in the early years of the first decade of the twenty-first century. That is not to say that such consultations did not take place, but that they were less formalised and open than subsequently. However, as noted above, differences in practices among Commission DGs persist, and informal consultations continue to be based on personal relationships, despite the standards for consultation put in place after the publication of the 2002 document (European Commission, 2002b).

In the Parliament, there is a system that provides accreditation for stakeholders to access the Parliament buildings and outlines a code of conduct applicable to them. As of January 2013, 5,522 individuals had signed this registry40 allowing access to the Parliament buildings. With much of the negotiation going on in the committees of the Parliament, they, and the rapporteurs on the particular dossiers, are often the focus of lobbying activities. As with the Commission’s DGs, however, there are differences among the committees with regard to their relations with external stakeholders (Tanasescu, 2009, p. 49). Much of the lobbying activity in the Parliament (and also with the Commission on a more informal basis) is based on personal relationships built up over time among lobby organisations and decision-makers (and their staff members; interviews 5, 6, 7, 8, 10, 11 & 26).

40 The list of accredited stakeholders (the “transparency register”) is available at: http://europa.eu/transparency-register/, accessed: 4 January 2013.

70 Conceptual framework

The Parliament’s code of conduct for lobbyists has also evolved over time. The Parliament was increasingly targeted by lobbyists, in line with the increased power assigned to the Parliament in the expansion of the co-decision procedure to more legislative areas in the Treaties (Hauser, 2011). The Parliament amended its rules of procedure governing the code of conduct of lobbyists in 1997 and worked with the Commission in the mid-2000s to develop the European Transparency Initiative, in order to ensure transparency on who was lobbying the Parliament (European Commission, 2006c; European Commission 2007d). Both stakeholders and MEPs acknowledged in interviews the increased level of lobbying of the Parliament over time, but described access to policymakers as easy and open (interviews 5, 6, 7, 25 & 26).

With the Council, there is less demand for external expertise, as member state representatives can possess or easily obtain the required information to be able to move forward on policy with their own national experts (Hauser, 2011, p. 698). Effective lobbying of the Council is rather about targeting specific member states, and is more likely to occur at the national level than at the European level. Interviewees described closed-door meetings and difficulties in accessing the Council and the Council working groups (interviews 5, 6, 7, 8, 10 & 11). But for a few exceptions, EU-level external stakeholder access is limited when it comes to the Council, in contrast to the Commission and Parliament, where access can be more easily attained. This may also reflect the fact that the policy process has usually passed through several phases (policy proposal from the Commission, negotiations in the Parliamentary committees and plenary first reading agreement in the Parliament) before the policy proposal is decided upon in Council’s first reading. Thus, many of the major issues may already have been discussed along the way (Hauser, 2011, p. 698). Additionally, negotiations in the Council are organised in a more intergovernmental fashion and negotiators may only have little room to deviate from a nationally agreed negotiating mandate. European-level external stakeholders may also face pragmatic choices within their own organisation on where to focus their lobbying activities (due to time, human and financial constraints, for example), with the Parliament and Commission providing more fruitful lobbying in many cases (interviews 5, 6, 7, 8, 10 & 11).

71 Conceptual framework

Finally, the third indicator for the measurement of the extent of CPI in the policy process includes the treatment of functional interactions with climate policy objectives in the policy discourse. Where actors in a policy sector recognise the potential interrelations between long-term (to 2050) climate policy objectives and their policy developments, and where these interrelations are continually recognised and referred to throughout the process, this can help lead to higher levels of CPI in the policy process (and, potentially also in the policy output). Measuring this indicator includes a document analysis of policy papers throughout the process to assess whether scientific arguments about the need to combat climate change are referred to throughout the development of a proposal. Scientific input (from officially recognised scientific bodies such as the IPCC) has been regarded by many studies on EPI as essential for ensuring environmental (or climate) objectives are integrated into non- environmental policy sectors (e.g. Jacob, Volkery, & Lenschow, 2008; OECD, 2002; Schout & Jordan, 2008a, 2008b). This document analysis is supplemented with interviewee opinions on the relative importance of climate change as a motivating factor in the policy development, and an assessment of media reports and literature around the policy process (demonstrating actor understanding of functional interrelations in the policy discussions). In this case, CPI is considered “very high” if a policy proposal and process recognises and articulates the functional interrelations between its sector and long-term climate policy, and aims to combat climate change through scientifically-grounded proposals within its sector (e.g. in line with IPCC recommendations for policy); “high” if combating climate change is a stated goal of the particular policy and remains so in the policy discussions throughout the negotiations, although with less ambitious proposed measures for combating climate change; “medium” if the policy is considered to interact with climate policy but without climate change motivating ambitious policy action or being a major issue in the policy discussions; “low” if climate change features only in an introductory manner in the policy proposals and is barely referred to throughout the policy negotiations; and “none/very low” if the interrelations with climate policy objectives are not recognised or are ignored in policy development. This five-fold scale ensures a sufficiently differentiated understanding of the level of recognition of functional interrelations by policymakers in the EU, without imposing too-rigid or too-flexible a scale. The five-fold scale also ensures similar understandings and measurements are used for the policy process and for each of the three indicators in the policy process.

72 Conceptual framework

The various steps in the scale were developed by establishing very high levels of CPI and moving down from these levels to very low levels.

Taking an aggregate measurement of these three indicators leads to a single score for the extent of CPI in the policy process. Table 3.1 summarises the operationalisation of CPI in the policy process.

Level: No/Very Very high or Low Medium High Low CPI complete CPI Indicator:

No Not Consulted Co-drafters and Leading on the involvement consulted, throughout the co-deciders policy proposal 1. Internal and no but process with other development and pro-climate visibility in the providing sectoral DGs, negotiations stakeholders policy process unsolicited committees and opinions Council formations

No pro- No or little Procedures allow Easy access to Invited and climate access to the access, but ease policymaking, opinions sought stakeholders policy of access strong pro- by policymakers, involved or process, pro- limited, pro- climate lobby access 2. External visible in the climate climate lobby with little guaranteed, pro- pro-climate policymaking stakeholders’ facing certain opposition from climate stakeholders process arguments opposition from other stakeholders’ receive other stakeholders arguments general stakeholders and/or receive general opposition and/or policymakers support policymakers

Functional Climate Functional Achieving Policy is interrelations policy interrelations climate policy developed in 3. Policy with climate objectives recognised and objectives is a order to achieve discourse: policy mentioned, considered stated goal of long-term climate recognition of objectives not yet throughout the policy and policy objectives interrelations recognised or interrelations policy process, continues to and such a with climate ignored not strongly yet without motivate policy standard remains policy featuring motivating development in place objectives throughout policy action throughout the throughout the policy process policy process process

Table 3.1: Indicators to measure the level of CPI in the policy process.

To come to an overall level for CPI in the policy process, I aggregate the three indicators on an equally weighted basis. No one indicator is assumed to play a greater role than the other. In practice, this may not be an entirely accurate reflection of the reality, and any occurrence of nuances found in the analysis of the case studies is discussed in the empirical chapters and in the conclusions. However, methodologically, an aggregated result has the benefit of providing a single overall

73 Conceptual framework score for the level of CPI in the policy process, to match the single overall score for the level of CPI in the policy output.

3.1.3.2 Measuring CPI in the policy output In terms of policy output, very high or complete levels of CPI will be achieved if policies are 100 per cent in line with established (and scientifically grounded) climate policy objectives. For climate policy, this means that policies must aim to achieve the long-term climate objective of reducing greenhouse gas (GHG) emissions by between 80 and 95 per cent by 2050 in the EU. This particular goal is supported by the scientific reports of the IPCC (IPCC, 2007) and also received political backing in the European Council meeting in October 2009 (European Council, 2009b). Therefore, not only is this objective scientifically grounded, but it is also a stated objective of the EU as a whole.41

With this standard in mind, an investigation of the levels of CPI in the policy output can establish the extent of the gap between the policy output of a particular policy measure and the ideal or very high level of CPI. For example, in renewable energy (RE) policy, a goal of achieving between 80 and 100 per cent share of renewables in EU energy consumption by 2050 can be considered as in line with 2050 climate policy objectives (EREC & Greenpeace, 2010; European Commission, 2011a; Heaps, Erickson, Kartha, & Kemp-Benedict, 2009; WWF, 2007). This policy goal is considered evidence of very high or complete CPI (see Table 3.2). However, if the RE policy output under examination achieves less than an 80 to 100 per cent share of RE in the EU by 2050, I can measure the distance between this very high standard and the actual output. Nonetheless, as most policies do not put in place specific measures to 2050 (but to, e.g. 2020 instead), it is rather by assessing the pathway that the policy output sets towards the 2050 goals that I can measure the level of CPI in the policy output. Taking again the example of RE policy, the EU’s 2020 goal of increasing the share of RE in the EU to 20 per cent is a policy output that may or may not put the EU on the pathway to achieving the 2050 climate goals. By examining the pathway that very high levels of CPI in the policy output would require, I can identify the sort of policy output for 2020 that would lead us to this level. The distance between this and

41 EPI is also a legal obligation enshrined in the Treaties (Article 11, TFEU, see Annex II), and with the TFEU, combating climate change is explicitly mentioned in Article 191(1) as an objective of environmental policy of the EU (see Annex II for the full text of Art. 191(1), TFEU).

74 Conceptual framework the actual policy output to 2020 represents the level of CPI found in EU RE policy to 2020.

Thus, on this basis, I strive to make an assessment of the level of CPI in the policy output by applying a scale ranging from very low over low, medium and high to very high/complete (see Table 3.2). Such a fivefold scale provides for a sufficient scaling of results. As this study is mostly qualitative – although there are some quantitative elements involved – the fivefold scale allows sufficient flexibility in the results to accommodate qualitative interpretations. Additionally, a threefold scale in this case would be too flexible, yet a greater number of ranges may run the risk of categorising results too rigidly (e.g. a seven-fold scale or more).

No/very Very high/ Low Medium High CPI in the low CPI complete policy output: 0-20% 21-40% 41-60% 61-80% 81-100%

Table 3.2: Scale establishing the extent of CPI in the policy output. For the purposes of establishing climate policy objectives for the case studies in later chapters, I refer specifically to a collection of EU-focused roadmaps and scenarios to 2050, or to certain global scenarios that outline actions for the EU also. These include the European Commission’s 2050 energy roadmap and its impact assessments (European Commission, 2011a, 2011e, 2011f, 2011g); the European Climate Foundation’s selection of scenarios to 2050 (ECF, 2010); the European Renewable Energy Council’s (EREC’s) “RE-thinking 2050”, and Greenpeace and EREC’s “energy [r]evolution” (EREC, 2010; EREC & Greenpeace, 2010); Eurelectric’s “power choices” (Eurelectric, 2010); Eurogas’ roadmap to 2050 (Eurogas, 2011); WWF’s “energy report” (WWF, 2011); Friends of the Earth’s report on “Europe’s share of the climate challenge” (Heaps, et al., 2009); and the IEA’s 2011 world energy outlook (IEA, 2011a, 2011b). Not all of these roadmaps and scenarios are directly comparable due to different sectors of focus, different objectives and methodologies. However, taking them together, a general picture of the possible (and desirable) goals for climate policy for 2050 can be drawn for each of the cases discussed in chapters four to six. In each case, reference is made to these studies and

75 Conceptual framework scenarios in order to establish the benchmark of very high levels of CPI against which the policy reality is measured.

3.2 Theoretical framework Now that I have discussed the operationalisation of CPI for measuring its extent in the EU energy policy sector’s process and output, it is time to outline the framework for explaining the extent of CPI found. To develop such an explanatory framework, I review the explanatory variables set out in general EPI literature and in several theories of European integration that, combined, can help provide a greater understanding of the EU policymaking process generally, and the extent of CPI specifically (Rosamond, 2007; Warleigh-Lack & Drachenberg, 2010; Wiener & Diez, 2009b). As such, I draw inspiration from broader theories of European integration to bring various explanatory variables highlighted in the EPI literature into a manageable framework. The project is not a theory-testing exercise but a study of the empirical reality in EU policymaking. Thus, the combination of several theoretical perspectives, as promoted notably by Wiener and Diez in their 2009 edited book on European integration theory, can help ensure as comprehensive an understanding of empirical results as possible (i.e. combining state- and process-centred perspectives, and not rejecting the role played by institutions and actors in EU policymaking). Such a conceptual combination of perspectives is increasingly promoted also in international relations theory as a pragmatic way of advancing knowledge (see, e.g., Friedrichs and Kratochwil, 2009).

3.2.1 Theoretical and explanatory frameworks in EPI literature EPI literature provides a disparate and long list of potential explanatory variables, depending largely on the conceptualisation of EPI. Whether EPI is viewed as an overarching legal perspective, a tool for efficient policymaking, a policy outcome or an aspirational goal, differing explanatory variables come to the fore. Few examples of truly comprehensive explanatory frameworks have evolved (Jordan & Lenschow, 2010; Persson, 2004; Dupont & Oberthür, 2012; Adelle & Russel, 2013).

I develop, therefore, a theoretical framework for understanding CPI in the EU based on the recurring explanatory variables recognised in EPI literature and derived from the crucial variables outlined in general theories of European integration, thus helping to explain CPI throughout the EU policymaking process and its output. Explanatory

76 Conceptual framework variables in EPI literature depend largely on the focus of analysis. Whether the focus of the analysis is on the policy process or output, or on the overarching principle of EPI, different explanatory variables have been emphasised. For example, research focusing on the policy process has employed an institutional perspective, emphasising coordination mechanisms (Jordan & Lenschow, 2008c), or a “policy learning” perspective highlighting the importance of policy framing (Nilsson, Eckerberg, Hagberg, Swartling, & Söderberg, 2007; Nilsson & Persson, 2003). A legal perspective has been employed to assess the overarching legal commitment to EPI (Nollkamper, 2002), while policy evaluation studies have attempted to assess EPI in the policy outcomes, emphasising resources, monitoring of policies and political commitment (EEA, 2005; Lafferty & Knudsen, 2007; OECD, 2002). Different theories have thus been used to understand the varied rates of EPI in different jurisdictions, including governance theories (von Homeyer, 2006), learning theories (Nilsson & Persson, 2003) and theories of bureaucratic politics (Lenschow, 2010; see Table 3.3 for an overview of different approaches for assessing EPI based on the analytical focus). Each of these perspectives helps to explain part of the EPI story, although most EPI research in the past has focused specifically on the policy process (Dupont, 2011; Adelle & Russel, 2013).

Approach(es) Variables Some key references EPI as Legal perspectives; EPI in constitution or Lafferty & Hovden, overarching normative (for EU) treaties; 2003; Nollkamper, principle perspectives. political commitment. 2002. Bureaucratic Procedural factors, such Jordan & Lenschow, politics; learning as decision-making 2008b; Nilsson & EPI as processes; procedures, Eckerberg, 2007; process governance consultations, Nilsson & Persson perspectives. organisational factors. 2003; von Homeyer, 2006. Governance Substantive factors, such EEA, 2005; OECD, perspectives; rational as end effects on the 2001; Lafferty & EPI as output decision making. state of the environment, Knudsen 2007; resources and funding, Mickwitz et al., 2009. monitoring mechanisms. Table 3.3: Summary of some of the main approaches and variables in EPI literature.

Source: adapted from Adelle & Russel 2013; EEA, 2005; Dupont, 2011; Persson, 2004. As this is a study on the extent of CPI in the EU, the explanatory variables highlighted in the EPI literature can be linked to the insights on EU policymaking found in general theories of European integration. Therefore, instead of following one of the

77 Conceptual framework many frameworks outlined above, the discussion below links the empirical findings in EPI literature to three European integration theoretical perspectives. These theories help focus the analysis on a certain number of explanatory variables, yet taken together, they can provide an encompassing framework for analysis that does not raise one perspective above another.

As a result of the review of EPI literature and theories of European integration, four main explanatory variables are defined and discussed below: functional interrelations; political commitment; institutional and policy context, which each explain CPI in both the policy process and output; and the process dimension, which explains CPI in the policy output only (Herodes, Adelle, & Pallemaerts, 2007; Jordan & Lenschow, 2008a; Lafferty & Hovden, 2003; Lenschow, 2002; Nilsson & Eckerberg, 2007; Persson, 2004; Dupont & Oberthür, 2012).

3.2.2 Theories of European integration General theories of European integration have been in development for as long as the European project has existed. Over the course of more than five decades, the number of theories has grown, while their scope has often narrowed. The so-called “grand theories” of European integration include neofunctionalism and liberal intergovernmentalism (earlier, neorealism). Other theoretical perspectives do not aim towards explaining or understanding the European integration project as a whole, but rather certain aspects of the European polity, policymaking processes or the politics of the EU (Wiener & Diez, 2009a, p. 245). In addition, new formulations of the “grand theory” of neofunctionalism have reduced its scope to a theory with only partial explanatory power (Stone Sweet & Sandholtz, 1998). For understanding EU governance and policymaking, the differing theoretical perspectives provide insights that emphasise different actors and processes from varied understandings of reality.

Rather than seeing the various theories as representing conflicting views and perspectives, I follow the understanding of Diez and Wiener that the theoretical approaches can be seen as “adding to a larger picture, without being combined into a single, grand theory” (2009, p. 17). In other words, in order to gain an understanding and explanation of (CPI in) EU policy processes and outputs, a combination of several theoretical perspectives is useful to aim for a comprehensive understanding. Therefore, it is my aim to combine the explanatory power of different theoretical

78 Conceptual framework perspectives to gain as full a picture as possible of CPI in the EU. I use a combination of three theories to inspire the explanatory framework. These are two so-called “grand” theories of European integration (neofunctionalism and liberal intergovernmentalism), and one more middle-ground theory (new institutionalism). These theories are particularly helpful for establishing a fuller explanation of the extent of CPI, due to their different areas of focus in the EU policymaking process.

These theories – neofunctionalism, liberal intergovernmentalism and new institutionalism – hold great potential for explanation as they focus on different aspects of the policymaking and decision-making processes, and place emphasis on the various roles of different actors. Liberal intergovernmentalism’s focus on political bargains and member states as the most important actors helps explain CPI with reference to the level of political commitment and leadership to climate change and CPI generally, and on the role of the Council and the European Council. Neofunctionalism provides explanations with reference to the involvement of multiple actors in the policymaking process, information exchange and the functional interlinkages between policy sectors. Finally, institutionalism can explain CPI by highlighting the importance of institutions, the policy and institutional context of the decision-making processes, and the legacy of previous policies and decisions.42 Although no one perspective, with its limited number of explanatory variables, seems to explain the level of CPI comprehensively, connections among the theoretical perspectives and the explanations they provide can be expected. Thus, since each of the theories focus on these different aspects of the policymaking process and with different assumptions about the drivers of European integration more generally, there is a case to be made in deploying these perspectives in a compatible way, without ascribing a hierarchy of importance to the various perspectives. In this way, the theories become tools for understanding the empirical results in a particular EU policy domain, as opposed to ideological frames for describing the overarching European integration project.

42 New institutionalism includes a branch called “sociological institutionalism”, which focuses on several elements of constructivist theory about the importance of ideas and socialisation. Neofunctionalism also includes the idea that elite socialisation is an important element of EU policy development. Although I do not explicitly focus on constructivist theories of European integration, some elements of constructivist theoretical perspectives may and do appear in the analytical framework.

79 Conceptual framework

3.2.2.1 Neofunctionalism Neofunctionalism dates from the 1950s and 1960s, when Ernst Haas and Leon Lindberg elaborated the theory to explain developments in the integration of the European Coal and Steel Community (Haas, 1958; Lindberg, 1963). The roots of this theoretical perspective can be found in functionalist and federalist thought, which emphasise integration through day-to-day decision-making, incremental changes and learning processes (Niemann & Schmitter, 2009). In its early development, neofunctionalism aimed towards grand theorising of European integration, by focusing on the nitty-gritty of day-to-day decision making that leads to “spillover” in other policy domains and increased integration (Diez & Wiener, 2009; Haas, 1958; Niemann & Schmitter, 2009; Strøby-Jensen, 2007; Wiener & Diez, 2009b). In the 1970s and later, however, the theory underwent reformulation and revision, with neofunctionalist scholars disagreeing on key issues, such as when integration is complete, and whether, or to what extent, policymakers’ loyalties move to the supranational level. Haas himself later criticised the theory for its apparent inability to explain periods of inaction in the EU integration process (Haas, 2001).

Neofunctionalist theorists consider integration to be a process, rather than agreeing on a specific outcome or end point – a process that continues throughout the everyday decision-making and the “spillover” between issue areas. Haas defined integration as a “process whereby political actors in several distinct national settings are persuaded to shift their loyalties, expectations and political activities towards a new centre, whose institutions possess or demand jurisdiction over the pre-existing national states” (Haas, 1958, p. 16). Emphasis is thus placed on the multiplicity of actors involved in EU decision-making processes, and – as a process-centred theory, rather than a state-centred theory – it differs in perspective to liberal intergovernmentalism (see section 3.2.2.2 below). Furthermore, neofunctionalists regard the role and influence of supranational organisations and the development of transnational interest groups as important to deepen and expand integration (Niemann & Schmitter, 2009).

Niemann and Schmitter (2009, pp. 48-49) outline five assumptions of neofunctionalism. First, actors are rational and self-interested (Haas, 1970, p. 627), but also have the capacity to learn and change their preferences. Second, institutions can develop in a way that increases their autonomy from their creators. Third, incremental day-to-day decision making impacts the integration process to a greater

80 Conceptual framework extent than grand political strategies or designs. Fourth, supranational decision- making is often about positive-sum games and compromises, rather than about veto- power. Fifth, interdependencies between economies tend to promote further integration.

Neofunctionalism’s core concept is “spillover”, which is generally sub-divided into three types: “political spillover”, “functional spillover” and “cultivated spillover” (Niemann & Schmitter, 2009, pp. 49-50). A process of “spillover” implies that decisions or actions taken in one particular policy sector lead to effects or pressures to move supranational policymaking to other policy areas (Pollack, 2010). This is described more specifically as functional spillover, referring to the organic process where cooperation in one policy area necessitates further cooperation in a (functionally) related area (Strøby-Jensen, 2007, p. 90). Political spillover describes deliberate political decisions taken to reach agreement in a variety of policy areas, and where supranational institutions (such as the Commission) and sub-national actors/interest groups push for further integration. Cultivated spillover describes the EU’s supranational institution members’ support for integration (Niemann & Schmitter, 2009, p. 50).

A second concept in neofunctionalist theory is that of “elite socialisation” – a concept describing the idea that the preferences and identities of people regularly involved in the policymaking process at the European level will gradually be “Europeanised”. This implies a shift in loyalties from the national to the supranational level, as outlined in Haas’ original definition of integration (Haas, 1958; Strøby-Jensen, 2007, p. 91).

Third, neofunctionalism presents the thesis that supranational interest groups develop over time and increase pressure for further integration (Strøby-Jensen, 2007, p. 92). Much like the concept of individual socialisation, interest groups are also expected to become more “European” over time, with increased involvement and interest in the supranational level.

Several theoretical criticisms have been levelled against neofunctionalism. First, critics have argued against the idea that actors in supranational institutions develop supranational loyalties, pointing instead to member states’ insistence on having their own national civil servants within these institutions (Strøby-Jensen, 2007, p. 93).

81 Conceptual framework

Second, neofunctionalism has been criticised for a lack of emphasis on the role of the member state as a driver or blocker of integration (Moravcsik, 1993). Third, the theory faced criticism over its normative aspirations as a model of integration, especially since it is regarded as “elitist” and disregarding of the need for popular legitimacy (Risse, 2005, p. 297). Finally, Haas has criticised the theory himself due to its difficulties in describing the evident lack of progress in European integration at certain periods in history (Haas, 2001). The development of new institutionalist theories (see section 3.2.2.3 below) aimed to respond to this criticism in some respects.

As neofunctionalism developed over time, scholars began to regard neofunctionalism not as a grand theory, but more as a theoretical perspective with partial explanatory power (Stone Sweet & Sandholtz, 1998). As mentioned above, during the 1970s, and during periods of low activity in the EU, early formulations of neofunctionalism seemed no longer to fit the empirical reality of European integration, and the theory was rarely used at this time because it could not explain the limited progress in European integration. The 1990s, however, saw a revival and a reformulation of neofunctionalism, ensuring its continued relevance. Stone Sweet and Sandholtz’s (1998) reformulation of neofunctionalism recognised the theory as one that partially explains European integration, disputing early theorists’ ambitions towards grand theorising. In conjunction with theories of institutionalism (see section 3.2.2.3 below), they presented a version of neofunctionalist theory that emphasised transnational exchange and transactions (ibid., p. 2). Thus, it became clear that neofunctionalism could continue to demonstrate valuable explanatory power, especially when combined with other theoretical perspectives for a more complete picture of integration.

3.2.2.2 Liberal intergovernmentalism Liberal intergovernmentalism aims to explain the large steps forward in integration (e.g. the major Treaty revisions), rather than the incremental steps in the day-to-day running of the EU (Cini, 2007; Diez & Wiener, 2009; Moravcsik & Schimmelfennig, 2009; Wiener & Diez, 2009b). It is a state-centred theory of European integration that has been developed, in particular, by Andrew Moravcsik since the early 1990s (Moravcsik, 1993; 1998). Its intellectual roots are found in traditional “intergovernmentalism” theory with “rationalist institutionalist” assumptions (Hoffmann, 1966; Moravcsik & Schimmelfennig, 2009, p. 67). Moravcsik developed

82 Conceptual framework this theoretical account of integration in response to the perceived weaknesses in the neofunctionalist account, especially during those periods in the history of the EU when little progress on integration was made (Pollack, 2010). As an account of the European integration process, it is praised for its clarity and parsimony (Risse-Kappen, 1996, p. 63) and it remains one of the most influential of the European integration theories (Cini, 2007, p. 109). Moravcsik describes liberal intergovernmentalism as the “baseline theory” of European and regional integration “against which other theories are often compared” (Moravcsik & Schimmelfennig, 2009, p. 67).

To summarise the main arguments and assumptions of liberal intergovernmentalism, Moravcsik points out that “European integration can best be explained as a series of rational choices made by national leaders” (Moravcsik, 1998, p. 18). Liberal intergovernmentalism is described as “an application of ‘rationalist institutionalism’” (Moravcsik & Schimmelfennig, 2009, p. 67), so its assumptions are based on models of rational actor decision making. The main assumptions underlying this theoretical perspective are, therefore, that states are the key actors in the EU, and that they are rational, unitary actors (ibid., p. 68). It is the member states’ preferences that guide and shape European integration in general.

The theory follows a three-step model, whereby, first, member state preferences are formulated (at the national level, not at the supranational level as neofunctionalist theory might suggest); second, member states bargain at EU-level; and, third, member states choose the institution to manage policy in a way that reflects member state commitments and recognition of the need for technical management (Moravcsik & Schimmelfennig, 2009; Pollack, 2010). These three factors (namely, state preferences; bargaining among states; and institutional choice) are thus the main explanatory variables posited by liberal intergovernmentalism to explain and understand advances in integration (Moravcsik, 1998; Moravcsik & Schimmelfennig, 2009). It is a state- centred theory, less interested in incremental change and process than neofunctionalism, and emphasising the central role of member states in any deepening of the integration project.

First, liberal intergovernmentalism aims to explain the motivations of states through an examination of their national preferences. Despite an acknowledgement that there is often a wide range of actors involved in policy formation, liberal

83 Conceptual framework intergovernmentalists still regard the state as a unitary actor since it is domestic politics and bargaining that generate the preferences of the state. Moravcsik also outlines that such preferences are subject to change over time, and that preferences vary among issue areas (Moravcsik & Schimmelfennig, 2009, p. 69).

Second, the interstate bargaining that takes place is explained following a rationalist perspective, and based on bargaining theory. In this context, liberal intergovernmentalists analyse the “transaction costs” (or the costs of intergovernmental negotiating) involved (ibid., p. 71). Bargaining power may flow from several factors, including the relative need among the parties to ensure supranational agreement in a particular sector (and the balance of incentives/threats in the bargaining process), and the information at hand to the various member states on the details and implications of a particular policy issue.

Third, liberal intergovernmentalists seek to explain states’ choice of institution for integration. In this respect, liberal intergovernmentalism follows on from “neoliberal institutionalism” (Keohane & Nye, 2011). Here, the emphasis is on states’ choice to delegate authority to supranational organisations in order to reduce future transaction costs, and to provide certainty about the preferences and behaviour of other states (Moravcsik & Schimmelfennig, 2009, p. 72).

There are four oft-cited criticisms of liberal intergovernmentalist theory in general. First, the theory is criticised for being too narrow in its scope to be considered a full- blown “grand” theory of European integration. Specifically, liberal intergovernmentalism cannot sufficiently explain the day-to-day politics and functioning of the EU (Cini, 2007, p. 112), but focuses rather on explaining the great treaty reform decisions. Yet, empirically, scholars note that integration continues (although less visibly and perhaps less swiftly) through incremental steps in the daily operations of the EU (ibid.).

A second criticism relates to Moravcsik’s narrow conception of the state. Critics argue that liberal intergovernmentalism ignores or underestimates the subtleties of domestic politics. In order to arrive at an informed understanding of state preferences, critics argue, a more in-depth analysis of the component parts of the state, and of domestic politics, is required (Cini, 2007, p. 113). Member state preferences can evolve as a result of a myriad of factors, including economics, party ideologies,

84 Conceptual framework domestic civil society interest groups, and state institutions (Pollack, 2010, p. 20). The formulation and preservation of member state preferences is thus more complex than liberal intergovernmentalism admits.

Third, the theory is criticised for its lack of recognition of the role of supranational actors in the European integration process. This includes the roles of supranational institutions such as the Commission and the European Court of Justice, and also the role of non-state “transnational” actors or interest groups (Cini, 2007; Stone Sweet & Sandholtz, 1998, p. 12). This links to the idea from constructivist critics of liberal intergovernmentalism that ideas, norms and rules from the EU level necessarily impact the formulation of preferences at the domestic level, with interest groups at supra- and subnational levels interacting (Pollack, 2010, p. 21).

Finally, liberal intergovernmentalism is often criticised by proponents of historical institutionalism for its lack of emphasis on the unintended or consequential subsequent effects of decisions (Moravcsik & Schimmelfennig, 2009, p. 75). Historical institutionalists also lament the fact that liberal intergovernmentalism seems not to take into account the difficulty of reversing decisions, once institutions are in place (ibid.; Pierson, 1998, pp. 28-30). This notion of institutional “stickiness” (Pollack, 1998) is discussed in more detail in the next section on new institutionalism.

3.2.2.3 New institutionalism Institutionalist analysis of the EU stems from a general re-insertion of “institutions” into many theoretical perspectives in international relations, including neorealism, Marxism and pluralism (Pollack, 2010). New institutionalist scholars contend that institutions “matter” and should not be absent from theories of international relations, governance or regional integration (Hall & Taylor, 1996; March & Olsen, 1989).

New institutionalist perspectives of European integration have developed since the 1980s and today can be divided into three versions: “rational institutionalism”, “sociological institutionalism” and “historical institutionalism” (Hall & Taylor, 1996; Pollack, 2009). These theoretical perspectives are, like neofunctionalism and unlike liberal intergovernmentalism, process-centred rather than state-centred theories (Jordan, Huitema, van Asselt, Rayner, & Berkhout, 2010; Wiener & Diez, 2009b).

85 Conceptual framework

Rational institutionalism highlights the logic of consequentialism – actors behaving strategically to try to realise their own preferences (Risse, 2009, p. 147) – and transaction costs. Here, institutions are either the independent variable explaining how institutions shape outcomes, or else they are the dependent variable created and managed by rational actors to perform certain functions. Liberal intergovernmentalism took over the concepts and assumptions of rational institutionalism as part of its three-step model to explain European integration, where the institutional choice is a rational choice made by member states (as the “rational actors”, see above).

Sociological institutionalism describes how institutions “constitute” actors and shape their views of the world, as individuals internalise the rules of behaviour (or culture) of institutions (Hall & Taylor, 1996, p. 948). Sociological institutionalism stemmed from sociology, rather than political science. Here, the emphasis is on a logic of appropriateness – i.e. actors’ behaviour is guided by rules, and the desire to do “the right thing” (Risse, 2009, p. 148), and thus “individuals simultaneously constitute themselves as social actors… and reinforce the convention to which they are adhering” (Hall & Taylor, 1996, p. 948). This is a constructivist understanding of the role of institutions (and the ideas, rules and norms associated with them) in preference formation and the shift of loyalty of actors working with and within institutions (thus linking to neofunctionalism with its emphasis on supranational loyalties developing over time, and disputing liberal intergovernmentalist assumptions that preference formation by member states is determined solely by domestic concerns). Sociological institutionalism highlights the values and ideas that permeate institutions (leading to persistence or change in institutional set-up/scope), rather than the rational objective of improved effectiveness as an explanatory variable for the persistence or change in institutional set-up/scope (ibid., p. 949).

Historical institutionalism, finally, focuses on the effects of institutions over time. In both historical and rational institutionalism, assumptions about the preferences of actors can be regarded as largely in line with rational choice theory (Pollack, 2009, p. 128). Rather than culture, socialisation, ideas and appropriateness of behaviour driving decisions (important factors for constructivists and sociological institutionalists), historical institutionalists consider the very structure of the institutional organisation as an important variable for explaining policymaking (Hall

86 Conceptual framework

& Taylor, 1996, p. 937). The main concepts in historical institutionalism include the so-called “lock-in” of decisions, and the fact that institutions are “sticky” and cannot be easily reformed (Peters, 2011). Therefrom comes the idea of “path dependence”, where future decisions are taken in the context of past policy trends and ensuing unintended consequences of previous decisions (Hall & Taylor, 1996, p. 938). Institutions are regarded as playing a key “role” in politics and policymaking, and the effect and unintended consequences of past decisions ought not to be underestimated. Historical institutionalism also points to the existence of the “joint-decision trap” (Scharpf, 1988; 2006), in which institutions remain in place, rigid and inflexible even in the face of a changing policy environment (Pierson, 1998, p. 45; Pollack, 2009). Historical institutionalism thus highlights the lack of emphasis on the “stickiness” of supranational institutions in liberal intergovernmentalism.

The main criticism that is levelled against new institutionalist theories is that they are regarded as mid-level theories, and therefore cannot encompass or explain the causes of integration in general or on a grand scale. Instead, they can simply examine the effects of institutions on EU politics and policymaking (Pollack, 2009, p. 142). As such, it is only by linking the mid-level theory and analysis to a broader, grand theory that a fuller explanation of integration might be possible. Both historical and rational institutionalist perspectives often face a second criticism. Sociological institutionalists and constructivists question the basic assumptions underlying this institutionalist perspective about the rational nature of actors and institutions (part of a greater debate between rationalist and constructivist scholars more generally; Pollack, 2010, p. 10). Such scholars argue in favour of an analysis taking account of the “constitutive” effects of EU institutions on the individuals interacting with them (Risse, 2009, p. 145).43

3.2.2.4 Application to research project Each of these three theoretical perspectives emphasise particular elements or parts of EU integration, governance and politics. Generally speaking, neofunctionalist scholars and institutionalist scholars consider that the European integration process can continue through further spillover among policy sectors and socialisation among

43 But see Dehousse & Thompson, 2012, who discuss the presence of “intergovernmentalists” in the Commission, and argue that EU-level socialisation may not lead to a shift in loyalties among policymakers.

87 Conceptual framework

European policymakers, elites and interest groups, and through “sticky” institutional set-ups. Liberal intergovernmentalists, and also rationalist institutionalists, view the possibility of continued European integration more sceptically, drawing evidence on the limits or lack of socialisation from cases such as the difficulty faced in ratifying the Constitutional Treaty (Pollack, 2010). Taken together, these perspectives can help understand the many interacting elements at work in the EU policymaking process. The advantage of drawing on a “mosaic” (Diez & Wiener, 2009, p. 19) of European integration theory for a study of CPI into the EU’s energy sector means that no one variable is emphasised to the exclusion of others. In this section, I discuss how variables derived from each of the theories of European integration, when linked with previous empirical evidence describing critical explanatory variables in EPI literature, help develop an encompassing explanatory framework for the study of CPI in the EU.

First, there is much potential in using neofunctionalism as one theoretical basis for a study of CPI. Functional spillover, firstly, can help explain the variation in the level of policy integration between policy areas according to the material interlinkages or “functional interrelations” of the policy sectors in question. The concepts of cultivated and political spillover, and of “elite socialisation”, secondly, can help explain the motivations of the member states and EU institutions – the Commission, Parliament and Council – for promoting (or blocking, ignoring) CPI. Neofunctionalism’s emphasis on supranational interest groups and on the multiplicity of actors in the decision-making process, thirdly, can help explain the (lack of) pressures in a given policy sector for integrating climate policy objectives. Thus, from a neofunctionalist perspective, emphasis is placed on the role and influence of non-state actors in the decision-making process. For these reasons, a neofunctionalist perspective could help explain the extent of CPI with a particular emphasis on the functional interrelations of the issue areas and the involvement of stakeholders in the policy process. A heightened role for stakeholders in the policy process has often been regarded as an enabler of higher levels of EPI (EEA, 2005; Herodes, Adelle, & Pallemaerts, 2007; Jordan & Lenschow, 2010).

Second, liberal intergovernmentalism could also hold some potential for explaining the level of CPI in the EU with its emphasis on the role of member states. In particular, this perspective can shed light on the extent of political will and commitment of member states to CPI (and thus an understanding of the role of the

88 Conceptual framework

Council and the European Council). The liberal intergovernmentalist emphasis on state actors, grand political decisions in the EU, and its focus on intergovernmental politics and member state preferences, point in the direction of grand political statements (from prime ministers, presidents and ministers) and their commitment to action. Different state preferences and their development over time can help us understand the level of political commitment at EU level and its evolution between 2000 and 2010.

Political commitment and leadership are generally considered very important for the establishment and development of policy integration (Jordan & Lenschow, 2008b; Lafferty & Hovden, 2003; Lenschow, 2002b; Persson, 2004). Even if other enabling variables are evident, without an overarching commitment and leadership from the political level, it has been argued that policy integration will not succeed or be sustained (Jordan, 2002, p. 35). Later literature also suggests that political commitment is just one part of the puzzle. A recent state of the art review of EPI found that political backing for EPI may exist in many jurisdictions (including in the EU), but without other enabling variables in place, this does not guarantee successful policy integration (Jordan & Lenschow, 2010, p. 147). Hence, a combination of explanatory variables is required for explaining levels of policy integration, of which one is political commitment.

Third, new institutionalist perspectives for the study of CPI can be useful especially due to their emphasis on policy pathways, and how institutional traditions and cultures impact the development of policy in the EU. The emphasis on how institutions matter in the policy process provides a complementary perspective to that presented in liberal intergovernmentalism and neofunctionalism. Thus, emphasis is placed on the importance of the EU’s supranational institutions, their decision-making procedures (and/or traditions), and past policy decisions in policy development. Additionally, new institutionalist perspectives can complement liberal intergovernmentalism’s focus on the political commitment of member states as an explanatory variable for CPI with an understanding that other institutions (the Parliament and Commission) could also demonstrate political commitment to CPI.

Both the institutional and policy context are highlighted as important variables for policy integration (EEA, 2005; OECD, 2002). This refers particularly to the

89 Conceptual framework governing architecture of the institutional framework for policymaking in the EU, and the influence of past policy and institutional decisions on present and future policymaking (Jordan & Lenschow, 2010; Pierson, 1998). For instance, we may hypothesise that an institutional set-up that requires unanimous decision-making may lead to less CPI than decision making by qualified majority voting (QMV), as integrating climate policy objectives into other sectoral policies necessitates change in policy, which can often be more easily achieved with majority voting rules. Where past policy actions have not achieved the desired results, this may allow for a window of opportunity to push for further policy developments. Furthermore, policymaking in the EU does not take place within a vacuum unconnected with external events, which may help push policy development.

3.2.3 Explanatory variables In summary, the explanatory variables for a study of CPI in the EU that can be derived from theories of European integration and that are highlighted in empirical studies on EPI are (1) functional interrelations; (2) political commitment; (3) institutional and policy context, each of which explain CPI in both the policy process and output; and (4) the process dimension to explain CPI in the policy output. Below I discuss each of these variables and establish whether they can explain CPI in the policy process, the policy output or both, how they can be identified, and finally, what position of hierarchy or weight might each explanatory variable hold.

3.2.3.1 Functional interrelations First, the nature of the functional interrelations between the two sectors being integrated holds potential as an explanatory variable. This variable relates to neofunctionalist theory that emphasises “spillover” as a driver of European integration (Niemann & Schmitter, 2009; Strøby-Jensen, 2007), where integration is regarded as a process that continues through the everyday decision making and the “spillover” between issue areas (Diez & Wiener, 2009; Haas, 1958; Niemann & Schmitter, 2009; Rosamond, 2005; Strøby-Jensen, 2007).

Functional interrelations can help explain the level of CPI found in the policy process (both at the policy proposal phase and with regard to amendments proposed throughout the policy negotiations) and in the policy output (in terms of the text of the final policy output and in terms of the specific aims of the final policy). As an

90 Conceptual framework explanatory variable, the nature of the functional interrelations between climate policy and the other policy sector in focus first forms the very basis of, and shapes any demand for, CPI. In the case of CPI in the EU’s energy sector, the nature of the functional interrelations among the sectors describes the interaction between energy policy objectives and the objectives of climate policy. At the same time, the kind and strength of this functional demand may help in understanding the actual level of CPI. To this end, it is useful to consider two different properties of functional interrelations. First, functional interrelations may be more or less direct (interrelations between policy objectives are obvious, closely linked and clear) or indirect (interrelations between policy objectives may be more obscure, distant or hidden by other objectives), which may have repercussions for the strength of the resulting demand for CPI. Second, the functional interrelations may be more or less synergistic or conflictual, which can be expected to affect the ease or difficulty of advancing CPI. For example, where interrelations between policy objectives are obvious, such as when policies promoting renewable energy (RE) generation positively affect policies aiming to reduce GHG emissions, the functional interrelations are direct and synergistic. Policies to promote the construction of importing natural gas infrastructure interrelates with climate policies to the extent that increasing consumption of natural gas is counter to long-term climate objectives, yet may be beneficial (by displacing coal) in the short-term. Additionally, such policies may divert funds and attention away from RE policies, and may focus on security of supply questions. In this second case, the functional interrelations are less clear, and are mainly conflictual.

Table 3.4 outlines a matrix of functional interrelations in relation to CPI. Where the functional interrelations between climate policy and the policy sector under development are direct and synergistic, higher levels of CPI could be expected, while indirect and conflictual functional interrelations could more likely block CPI from occurring. Thus, we can describe situations where the functional interrelations between a policy sector and long-term climate policy objectives are both direct and synergistic as providing most favourable conditions for the recognition, articulation and advancement of CPI in the policy process and in the policy output.

91 Conceptual framework

Direct Indirect

Synergistic + + + -

Conflictual - + - -

Table 3.4: The nature of functional interrelations and their potential effect on CPI.

3.2.3.2 Political commitment Second, the political commitment demonstrated by the EU’s leaders (including, but not only, heads of state and government) could be considered a core variable for explaining CPI in both the policy process and output. Drawing on liberal intergovernmentalism, political commitment44 can be found in the preferences of the member states. Thus, evidence for political commitment will surface in the conclusions of the Council (most pertinently for this research project, the energy and environmental Council conclusions) and the European Council. New institutionalist perspectives direct us also to consider the political commitment demonstrated by other EU institutions – thus the policy proposals and statements of the Commission and the proposed amendments and statements of the Parliament may also provide an indication of the level of political commitment overall in the EU. Political commitment is thus recognised in the stated objectives of the political members of the EU, and in follow-up, the decisions taken to see these objectives materialise.

With regard to CPI, two aspects of political commitment appear to be particularly relevant. First, a general political commitment to and support for climate policy objectives plays an important overarching role when it comes to CPI. Second, the political commitment to climate policy integration into the specific policy sector under investigation, based on an acknowledgement of the relevance of any overall climate policy objectives for the sector, should be important for the prospects of CPI actually being advanced. Political commitment to climate policy and CPI from the outset of policymaking can affect the extent to which the functional interrelations between the policy sector and climate policy are emphasised in the policy proposal and throughout the policy negotiations. High levels of political commitment among

44Political commitment can also be described as “political will” (to combat climate change/advance CPI).

92 Conceptual framework

EU policymakers may push CPI to the extent that the final result is a policy output that is far-reaching in terms of achieving climate objectives. The political commitment can thus explain the level of CPI in the policy output as well as in the policy process.

Here, I measure political commitment to climate policy and to CPI with reference to the public statements of EU member states, leaders and institutions, and assess the existence of concrete follow-up on the expressed commitment (in terms of legislation, for example). Table 3.5 outlines the scale for measuring political commitment. I use a three-fold scale in this case, as assessing the EU’s political commitment to combating climate change and to advancing CPI is necessarily a highly qualitative task. The indicators used to identify the different levels of CPI vary according to the evidence of commitment in political statements, followed up with concrete action. In this case, a three-fold scale is sufficient to encompass the variation.

Low Medium High

Expressed commitment No evidence of Expressed to climate change or Measuring commitment to commitment to CPI in statements, and political climate change or climate change or actions to follow commitment CPI in statements CPI in statements through on the commitment

Table 3.5: Scale measuring political commitment to, first, combating climate change, and, second, advancing CPI. 3.2.3.3 Institutional and policy context Third, a new institutionalist perspective leads us to pay particular attention to the institutional and policy context for policy integration in both the policy process and the policy output (Hall & Taylor, 1996; Pierson, 1998; Pollack, 2009).

The institutional context refers, particularly, to the European institutional set-up, including the decision-making procedures in the EU’s supranational institutions. The policy context can refer to the path dependency that is created for present and future policies, due to past policy and institutional decisions (Jordan & Lenschow, 2010; Pierson, 1998), as well as external policy factors (e.g. international negotiations on climate change) that may influence policy development in the EU. EU integration overall, such as enlargement, may affect the dynamic of policymaking. Actors may learn from past policy failures or successes and previous decisions may create or

93 Conceptual framework undermine a dynamic that facilitates a push for policy change (thus potentially allowing windows of opportunity for more or less CPI) throughout the policy process. Institutionally arranged decision-making procedures may also affect the extent of CPI in the policy output. Decisions by qualified majority, for example, may be assumed to facilitate policy change towards CPI as compared with a consensus rule.45

While not particularly emphasised in the theories of European integration outlined above, the external political and policy context nevertheless may play a role in facilitating policy development that may be more or less conducive to enhancing CPI. We can expect certain external “shocks” (such as energy or financial crises) or events (such as international climate negotiations; Nohrstedt, 2006; Wettestad, 2005) to impact the policymaking process (and policy priorities) within the EU, by opening windows of opportunity for policy development (Kingdon, 2003).

To assess this variable I analyse policy documents to discover the institutional set-up in the policy sector in question. Here, I examine what legal basis is chosen for the proposal, what this choice means for the institutional negotiations, and whether the policy is negotiated under the co-decision procedure (and thus whether QMV rules apply). Next, I look to the policy context, which involves assessing the timing and sequencing of the policy negotiations, and whether or not the external political context can be considered to play a role in policy development. This variable explains levels of CPI in that the more emphasis on combating climate change, both in the institutional context (e.g. if environmental institutional actors are playing a leading role in the policy process) and the external political context (e.g. if policy is negotiated on time for international climate negotiations), the greater the expectations to find higher levels of CPI. Thus, the institutional and policy contexts can be said to be either more or less favourable for CPI, and especially with regard to the levels of CPI in the policy process. The institutional and policy context is also valuable for understanding the levels of CPI in the policy output, as certain of these contextual issues may swing the balance in the final policy decision.46

45 Procedural participation rights, transparency requirements and other procedural rights that are usually emphasised by new institutionalism are incorporated in the explanatory variable covering the role of the policy process to explain the policy output (section 3.2.3.4).

46 An implicit understanding is also that the policy output generally follows from the policy process. It would be surprising (but not impossible) to find a policy output that was not influenced by the negotiations and context that played out during the policy process.

94 Conceptual framework

3.2.3.4 Process dimension Finally, I expect the level of CPI in the policy process to play an explanatory role for understanding the extent of CPI in the policy output. The higher the level of CPI in the policy process, the more likely higher levels of CPI will also be found in the policy output (Briassoulis, 2005). This explanatory variable relates to both new institutionalist and neofunctionalist theoretical perspectives, which emphasise the importance of multiple actors in the policymaking processes (see above). Whether procedures are in place in the policy process to allow access for environmental and climate actors can affect the level of CPI in the policy process, which in turn may explain the extent of CPI in the policy output. One assumption here is that where CPI does not feature highly in the policy process it would be surprising (but not entirely impossible) to then find high levels of CPI in the policy output.

Furthermore, for functional interrelations and their properties to have political effect, they need to be part of the political discourse and realised in the policy process. This final aspect also requires attention in the analysis. For the purposes of a study on CPI, an analysis of the functional interrelations of policy sectors under examination can be usefully complemented with a document analysis, and with interviewee responses, to establish whether it is acknowledged in the various sectors that such functional interrelations exist and can affect the policy objectives of that sector. Thus, breaking down the extent of CPI in the policy process could help provide more understanding of the level of CPI in the policy output.47

3.2.3.5 Hierarchy among the explanatory variables These explanatory variables may interact with each other to provide a clearer explanation or understanding of CPI and some may prove more important in the explanation than others. A first hypothesis about the hierarchy or weight of these variables, based on the literature on EPI, is that political commitment to climate policy and to climate policy integration will prove crucial for the emergence of CPI in EU energy policy. However, I would suggest that political commitment is an insufficient explanatory variable. It cannot, by itself, explain levels of CPI in EU energy policy process and output. A more nuanced hypothesis would suggest that a

47 For more on details on establishing the level of CPI in the policy process, and, thus, helping to explain the level of CPI in the policy output, see section 3.1.3 above.

95 Conceptual framework combination of variables is required for CPI to manifest itself in the policy process and output, of which political commitment is one important one (Jordan & Lenschow, 2010; Adelle & Russel, 2013).

The second hypothesis then stems from this first one, namely that it is in fact the recognition of functional interrelations that is a baseline and crucial condition for CPI. In this sense, the recognition and articulation of functional interrelations could be considered a first-order explanatory variable. This precludes evidence of CPI in the policy process (which can affect the extent of CPI in the policy output as a result). Where a policy sector interrelates directly and synergistically with climate policy, political commitment to CPI and high levels of CPI in the policy process could more likely be manifested. The opposite (indirect and conflictual functional interrelations combined with low political commitment and low CPI in the policy process) may also occur. Therefore, we can consider that functional interrelations must crucially be recognised in the policy process. The level of CPI in the policy process can be considered a variable that is contingent on this first condition. Without the recognition of the functional interrelations with climate policy in the policy process, CPI in the policy process could be low. Following from this low level of CPI in the policy process, low levels of CPI in the output may also be expected.

Thus, combining hypotheses one and two, I suggest that the recognition of functional interrelations is a crucial enabler of CPI, and combined with political commitment to climate policy and CPI, higher levels of CPI can be expected. The remaining explanatory variables, namely the institutional and policy context, and the level of CPI in the policy process, can shift the balance of CPI to lower or higher levels, but only if the functional interrelations had already been recognised from the early stages of the policy process. The external policy context and the institutional context can be said, thus, to represent background contextual factors that can interact with any of the remaining variables in such a way that they can have more or less effect on CPI. For example, the economic and financial crises that first gripped the world and Europe in 2008 may result in a weakening of political commitment to climate policy objectives, which in turn can explain potentially lower levels of CPI. The gas supply crises to Eastern Europe in 2006 and 2009 may hide the functional interrelations between climate policy objectives and increased natural gas infrastructure projects. The dominance of traditional energy company lobbies in energy policymaking overall

96 Conceptual framework may affect the access of pro-climate stakeholders to the policymaking processes, and thus affect the level of CPI in the policy process. Therefore, the four explanatory variables are not expected to hold equal weight in all case studies, and their relative importance may depend on the existence, first and foremost, of the recognition of the functional interrelations.

3.3 Conclusions This chapter outlined the conceptual and analytical frameworks applied to the empirical study of CPI into the EU’s energy sector. Conceptualising CPI as a matter of degree, where “principled priority” for climate objectives is considered a methodological tool for identifying very high or complete levels of CPI provides the advantage that the analysis can encompass all levels of CPI, from weak to strong, and enhance comparison with other similar studies. Operationalising this concept requires the measurement of CPI in both the policy output and the policy process. With regard to the policy output, a complete level of CPI would mean that the policy sector objectives are 100 per cent in line with (scientifically grounded) climate objectives – such as the long-term objective to reduce GHG emissions in the EU by between 80 and 95 per cent by 2050 (based on IPCC recommendations for developed country action to ensure that global temperature increase does not exceed 2° Celsius). For the policy process, procedures guaranteeing the access to the policymaking process for internal (within the EU institutions) and external pro-climate policy stakeholders, as well as the recognition and articulation of the functional interrelations with long-term climate policy objectives in the policy process, can enable higher levels of CPI.

Overall, the review of literature on EPI and of general theories of European integration allows for the identification of a limited number of core variables that can be systematically employed to explain varying levels of policy integration in general, and CPI in particular, across a number of cases. Taken together, these core variables and their identified components provide a differentiated but manageable framework for the systematic exploration and explanation of CPI into other policies and sub- policies. This explanatory framework highlights four main explanatory variables: the nature of functional interrelations; political commitment to climate policy and to CPI; and the institutional and policy context, which each explain CPI in both the policy

97 Conceptual framework

process and output; and, the process dimension, which can explain CPI in the policy output only. For a summary of the explanatory framework, see Table 3.6.

Explains Theoretical CPI in Variable Elements Operationalisation perspectives process or output?

Direct or Nature and type of functional Functional Process and indirect; Neofunctionalism interrelations between interrelations output Synergetic or two sectoral policy conflictual objectives

Council and Overarching to European Council Liberal climate policy conclusions; Political Process and intergovernmentalism objectives; statements and commitment output & new institutionalism To CPI, in follow-up of EU particular institutions (including Com and EP)

Institutional context (procedures); Legal basis; Institutional New institutionalism Process and past policy policymaking and policy & neofunctionalism output decisions; procedure; timing of context external policy process shocks or events

Pro-climate stakeholder and Extent of CPI advocate involvement in process; Process Neofunctionalism & and access to the Output only discourse on dimension new institutionalism policy process; climate policy acknowledgement of in process functional interrelations

Table 3.6: Summary of explanatory framework, as derived from theories of European integration and EPI literature. The next three chapters will apply this framework to the empirical examination of cases of energy policy in the EU. Chapter four examines CPI in the EU’s renewable energy policy, chapter five explores CPI in the EU’s energy performance of buildings policies, and chapter six analyses CPI in the EU’s policies to support external natural gas import infrastructure. Each case is examined over time: from 2000 to 2010.

CPI into EU renewable energy policy

This chapter is the first of three chapters each exploring one empirical case study. The case studies aim to measure the levels of CPI in the policy process and output of selected EU energy policies. As described in chapter one, the cases selected include one policy area on EU energy production (renewable energy policy); one policy on EU energy consumption (energy performance of buildings policy, chapter five); and one policy on external energy imports (external natural gas infrastructure, chapter six). Here, I discuss the case of renewable energy (RE) policy development in the EU, and examine and analyse the level of CPI in both the policy process and policy output for RE policy between 2000 and 2010. I therefore focus on the 2001 renewable electricity Directive (RES-E Directive 2001/77/EC) and the 2009 renewable energy Directive (RE Directive 2009/28/EC).

The chapter opens with a discussion of the development of RE policy, technology development and deployment in the EU generally. In section two, I outline benchmarks for very high or complete levels of CPI in the policy process and output of RE policy. The very high levels of CPI in the policy process are measured based on indicators outlined in Table 3.1. For the policy output, benchmark for very high levels of CPI is derived from roadmaps and scenarios for achieving the long-term climate policy objective of reducing GHG emissions in the EU by between 80 and 95 per cent by 2050. The benchmarks serve to analyse the level of CPI to be found in the policy process and output of the 2001 RES-E and the 2009 RE Directives. Third, I measure the actual levels of CPI in the policy process and output of both the 2001 RES-E Directive and the 2009 RE Directive. In the fourth section, I explain the levels of CPI found, using the explanatory variables described in chapter three. The final section summarises the main points of the analysis and concludes the chapter.

4.1 Renewable energy in the EU The energy sector (including energy production, transportation and consumption) is the highest fossil fuel consuming sector of society, and accounts for about 80 per cent of GHG emissions in the EU (CANEurope, 2011; European Commission, 2011a),

99 CPI into EU renewable energy policy with the remaining emissions coming from agricultural, waste and industrial processes.48 It is clear (and this is acknowledged by the Commission; European Commission, 2011b) that any attempts to mitigate climate change must include serious action to reduce GHG emissions in the energy sector. When it comes to mitigating climate change, switching away from fossil fuels and moving to renewable energies is one major strategy that can be employed: promoting RE displaces fossil fuel energy sources and directly reduces GHG emissions.49 Mitigating climate change is thus one major rationale for developing RE policy, along with aims to achieve the energy security objectives of reducing dependence on imports of fossil fuels, and especially of oil (Boasson & Wettestad, 2013; Howes, 2010, p. 17). However, historically, not much attention was paid to RE policy at the EU level. It did not feature strongly on the EU agenda until the late 1980s saw the rise of climate change as a global political issue.

4.1.1 Early RE policy development In the wake of the oil crises in the 1970s, some member states began to develop policies to promote RE development (see also chapter two). The first oil crisis came in 1973, following the Arab-Israeli war. This crisis highlighted the high levels of (and growing) dependence of the EU on imports of fossil fuels. The second oil crisis hit in 1979 after the Iranian revolution and continued during the subsequent Iran-Iraq war. This further imbued a sense of vulnerability among member states. The result was a search for alternative energy sources that could increase energy independence (Hildingsson, Stripple, & Jordan, 2010, p. 105). The motivations for developing renewable sources of energy thus stemmed rather from energy security than environmental or climate concerns.

However, these early policy developments took place at the level of some member states, and not at the EU level. Denmark, Sweden and Germany, in particular, first developed national RE policies in the 1970s (Nilsson, 2011, p. 113). These countries opted for RE policy development in response to the oil crises, although policies to

48 See www.eea.europa.eu for updated data of the share of GHG emissions per sector in the EU, accessed: 13 January 2013.

49 The majority, but not all renewable sources of energy, have beneficial effects on the reduction of GHG emissions. Biofuels are a controversial RE source due to potential effects due to land-use change associated with their production, and difficulties in calculating the GHG emission reduction potential of the various types or biofuels (Bowyer, 2010; Eickhout et al., 2008). In this chapter, I do not focus on biofuels specifically.

100 CPI into EU renewable energy policy promote any type of domestic energy production (more fossil fuels, nuclear, for example) could have achieved the same energy security objectives. At the EU level, RE policy did not really come onto the agenda until the 1980s.

As environmental concerns began to rise on the agenda throughout the 1980s and 1990s (Lenschow, 2002a; see also chapter two), climate change became a driver of RE policy on the EU level in addition to energy security concerns. By the 1980s, several member states had put in place their own instruments and support mechanisms for RE, resulting in a myriad different approaches across the EU (Lauber, 2001). The member states also favoured different renewable technologies (e.g. Denmark supported wind, Sweden supported bioenergy crops; Nilsson, 2011, p. 113).

Policy development at the EU level emerged partly as a response to these differentiated measures at national level. EU policy aimed to improve coordination in achieving an overall increase in the share of renewables in the energy mix of the EU, thus achieving both energy security and environmental objectives. EU RE policy, however, also emerged in response to the push to develop a liberalised internal energy market (Hildingsson, et al., 2010, pp. 103 & 106). With the drive towards market liberalisation, which gained momentum after the adoption of the Single European Act (1987) in the late 1980s and 1990s, issues arose related to the perceived unequal playing field among RE producers and conventional energy producers (Jansen & Uyterlinde, 2004, p. 93). There was thus a perceived need (especially in the Commission) for further liberalisation and the development of EU-level harmonised RE policies (Boasson & Wettestad, 2013).

With these motivations, RE therefore began to emerge on the EU level as an issue in the late 1980s, although it was not until the 1990s that the first signs of concrete policy emerged. The Council outlined in its 1986 resolution on Community energy policy objectives that the promotion of RE was one of the EU’s energy policy objectives (Council of the European Union, 1986; European Commission, 1997b, p. 6). Yet it was only in 1993 that the EU agreed to implement ALTENER, which was the first EU-wide initiative to promote RE. The aim of ALTENER was to increase the share of renewables to eight per cent (doubling the share); to treble the share of renewables in electricity generation, and to ensure five per cent share of biofuels in transport, all by 2005 (European Commission, 1992b, p. 24). In addition, the

101 CPI into EU renewable energy policy

Commission stated the rationale for EU-level action on RE was the role of the EU in coordinating national efforts, and “ensuring their convergence towards common objectives”. Setting quantified objectives at EU level, the Commission argued, gives clear indications to consumers, producers and investors (European Commission, 1992b, p. 21). It is clear that, at this stage, climate concerns began to play a more significant role in the justification for RE policy development. As the Commission stated, the EU wished to show that “the Community and its member states are determined to make a significant contribution to protecting the environment, and in particular reducing CO2 emissions, by exploiting RE sources” (European Commission, 1992b, p. 21).

With the failure of the EU’s CO2/energy tax proposal in the 1990s, ALTENER was one of the few internal policies that the EU could claim as having a climate impact (see also chapter two). While ALTENER was a first step towards developing real EU RE policy, the instrument itself was weak, mostly due to a lack of allocated funds. In 1996, the Commission followed up with a green paper on renewable energy (European Commission, 1996), and in 1997 with a white paper (European Commission, 1997b). An EU RE policy was slowly developing and gaining in strength throughout the 1990s.

4.1.2 Development in RE generation in the EU Renewable sources of energy are defined as renewable non-fossil energy sources that are “replenished by natural sources at a rate that equals or exceeds” their rate of use (Moomaw et al., 2011, p. 3). These are identified in EU legislation as including wind, solar, geothermal, wave, tidal, hydropower, biomass, landfill gas, sewage treatment plant gas and biogases (Directive 2001/77/EC Article 2(a); Directive 2009/28/EC Article 2(a)).

According to data from the European Environment Agency (EEA), the share of RE in the EU27’s final energy consumption had increased from 6 per cent in 1990 to 11.7 per cent in 2009. This included 13.1 per cent of RE in final heat consumption (compared to 6.6 per cent in 1990), 19.6 per cent of electricity consumption (11.8 per cent in 1990) and 4.1 per cent of transport fuel consumption (0.02 per cent in 1990).50

50 See www.eea.europa.eu database. The share of RE in final energy consumption is coded by the EEA as indicator “ENER28”, accessed: 8 October 2012.

102 CPI into EU renewable energy policy

Figures from Eurostat for 2010 showed another increase to 12.5 per cent share of RE in final energy consumption. Clearly, progress had been made over the two decades from 1990. The figures for 2009 and 2010 are likely to have been affected by the economic crisis that began in autumn of 2008. As overall levels of energy consumption fell in 2009 as a result of the crisis, the share of RE in final energy consumption could increase (ENDS Europe, 2011), although absolute levels of RE generation may not have increased to such a large extent. Figure 4.1 shows the development in the share of RE in final energy consumption between 2004 and 2010 (the timeframe for which data on this indicator has been recorded by Eurostat).

Figure 4.1: Percentage share of renewable energy in final energy consumption in the EU27, 2004- 2010.

Source: Eurostat data, epp.eurostat.ec.europa.eu.

Before 2004, the EEA highlights limited progress on improving the share of RE. The main improvements in the share of RE in final energy consumption happened between 2005 and 2009. In 2004, the share of RE stood at just 8.1 per cent (compared to 6 per cent in 1990). This represents a very small increase between 1990 and 2004, as opposed to the increase from 8.1 per cent share in 2004 to 12.5 per cent in 2010. One conclusion from these figures could be that RE policies were having a concrete impact on the development of RE in the EU in the late 2000s. Another conclusion is that the economic crisis from 2008 artificially inflates the share of RE in the EU as the overall level of energy consumption fell. However, it is true that the installation of RE technology in the EU has increased (EEA, 2011).

103 CPI into EU renewable energy policy

The development in RE was not universal among all member states. In 2010, Sweden had the greatest share of RE in its final energy consumption (nearly 48 per cent, up from nearly 39 per cent in 2004), while Malta had not yet succeeded in developing much RE generation (0.2 per cent of its final energy consumption came from RE in 2010, compared to no RE in 2004). Denmark also succeeded in increasing its RE share from about 15 per cent in 2004 to 22 per cent in 2010. Austria made a similar increase, from a nearly 23 per cent share in 2004 to about 30 per cent in 2010. Although the shares of RE vary among the member states, nearly every member state nevertheless increased its percentage share of RE in final energy consumption between 2004 and 2010 (see Table 4.1).51 The sources of RE also continued to vary among the member states.

51 Latvia is an exception: its share of RE in final energy consumption stood at 32.8 per cent in 2004 and 32.5 per cent in 2010.

104 CPI into EU renewable energy policy

2004 2010 2020 target EU27 7.9 12.1 20 Austria 22.8 30.4 34 Belgium 1.9 4 13 Bulgaria 9.2 13.4 16 Cyprus 2.7 4.6 13 Czech Republic 5.9 8.4 13 Denmark 14.9 22 30 Estonia 18.4 24.6 25 Finland 29 31 38 France 9.1 11.4 23 Germany 4.8 10.7 18 Greece 7.1 9.2 18 Hungary 4.4 7.6 13 Ireland 2.4 5.6 16 Italy 4.9 9.8 17 Latvia 32.8 32.5 40 Lithuania 17.2 19.8 23 Luxembourg 0.9 2.9 11 Malta 0 0.2 10 Netherlands 1.8 3.3 14 Poland 7 9.3 15 Portugal 19.3 22.7 31 Romania 17 22.9 24 Slovenia 16.1 19.6 25 Slovakia 6.7 8.5 14 Spain 8.1 13.8 20 Sweden 38.3 47.9 49 United Kingdom 1.1 3.3 15 Table 4.1: Percentage share of RE in final energy consumption in EU27 and in each member state in 2004 and 2010, with 2020 target.

Source: Eurostat data, epp.eurostat.ec.europa.eu.

105 CPI into EU renewable energy policy

4.2 Measuring CPI in EU RE policy In this chapter, I measure and analyse the level of CPI in the policy output and process of the 2001 RES-E and the 2009 RE Directives. In order to do so, it is necessary first to describe the benchmark against which I measure the levels of CPI.

4.2.1 Benchmark for measuring CPI in the policy process As discussed in chapter three, in which I described the conceptualisation and operationalisation of CPI, measuring the level of CPI in the policy process does not involve a single benchmark for a very high or complete level of CPI in the process. Rather, as outlined in Table 3.1, and based on a review of literature on EPI especially, it is possible to highlight a number of indicators for measuring the level of CPI in the policy process. In this study, I measure the level of CPI with reference to the involvement of internal pro-climate stakeholders (namely, those policymakers within the EU institutions that hold a pro-climate stance, such as DG Environment, the environment committee (ENVI) in the Parliament, and the environmental Council formation); the involvement of external pro-climate stakeholders (namely, stakeholders outside the EU policymaking institutions but who attempt to influence the policy process through consultation and lobbying and who have a pro-climate stance, such as environmental NGOs or the RE industry); and whether policymakers recognise (and articulate) the functional interrelations between long-term climate policy objectives and (in this case) RE policy during the process.

In the case of RE policy developed in the EU between 2000 and 2010, very high levels of CPI in the policy process would therefore see DG Environment as the lead drafters of the policy proposals in the Commission, with the ENVI committee drafting the report for the Parliament’s readings, and the environment Council formation negotiating and agreeing the Council’s position on the policy proposal. External pro- climate stakeholders would have easy access to provide input to the policy process and pro-climate arguments would receive general backing. Additionally, the motivation of policymakers to advance RE policy would be to achieve the long-term climate policy objectives. In this respect, the recognition of the functional interrelations between the policy areas in the process should help push for more ambitious policy measures to improve the share of RE in EU final energy

106 CPI into EU renewable energy policy consumption (given that increases in the shares of most types of RE can directly displace fossil fuel-generated energy, thus reducing GHG emissions).

4.2.2 Benchmark for measuring CPI in the policy output Establishing the level of CPI in the policy output of EU RE policy involves outlining a benchmark for expected very high or complete levels of CPI in the final agreed policy. The distance between the policy output and this very high or complete level of CPI shows the level of CPI in the policy’s actual output. To establish the benchmark I discuss what type of policy output for RE policy would be in harmony with 2050 climate policy goals of reducing GHG emissions in the EU by between 85 and 90 per cent compared to 1990 levels.

In this policy sector, very high or complete levels of CPI would imply ambitious policy to increase the share of RE in final energy consumption in the EU. RE policy goals to increase the share of (most sources of) RE also achieve climate policy objectives by displacing fossil fuels, and thus reducing GHG emissions. Therefore, it can generally be said that the more ambitious RE policy is, the better this is for achieving climate policy goals. When RE policy output includes ambitious and stringent targets for increasing the share of RE in final energy consumption, then we could expect higher levels of CPI.

Many scenarios outlining the road to decarbonisation by 2050 highlight the significant role to be played by RE in the energy mix to achieve the climate goals (e.g. ECF, 2010; EREC & Greenpeace, 2010; European Commission, 2011g). Although some scenarios for decarbonisation by 2050 aim for close to 100 per cent of RE by 2050 (EREC & Greenpeace, 2010; Heaps, Erickson, Kartha, & Kemp-Benedict, 2009; PricewaterhouseCoopers, 2010; WWF, 2011), other analyses include a range of solutions including carbon capture and storage (CCS) technologies, and nuclear energy in their assessments (see ECF, 2010; European Commission, 2011g). Depending on the assumptions regarding CCS and nuclear energy, most scenarios nevertheless imply an RE share of between 55 and 100 per cent by 2050 as required for decarbonisation (ECF, 2010; EREC & Greenpeace, 2010; European Commission, 2011g; Heaps, et al., 2009; WWF, 2011). With CCS technologies commercially unviable in 2013 (Reichardt, Pfluger, Schleich, & Marth, 2012), and nuclear energy

107 CPI into EU renewable energy policy continuously facing public opposition for both environmental and safety reasons,52 I follow here the arguments outlined in the decarbonisation scenarios that argue that a high proportion of RE is required in the overall energy mix for 2050. Scenarios that limit or exclude both nuclear and CCS technology suggest that close to 100 per cent of our energy demands can be supplied by RE sources. Thus, taking the top ranges outlined in studies on decarbonisation to 2050, that limit (or exclude) the role of nuclear energy and CCS, suggests between 80 and 100 per cent of RE share by 2050.

Figure 4.2 shows a linear trajectory for CPI towards 80 to 100 per cent share of RE in the EU between 2000 and 2050.53 This is compared to business-as-usual (BAU) scenarios from the Commission’s 2007 renewable energy roadmap and from the Commission’s 2011 energy roadmap (European Commission, 2007a, 2011g). A share of RE of between 80 to 100 per cent by 2050 implies an increase by about 7 to 9 percentage points every five years from 2000. In 2000, the share of RE stood at nearly 8 per cent (European Commission, 2011g), and this increased by just one percentage point to nearly 9 per cent in 2005 (ibid.; EEA, 2008, p. 44). The Commission’s BAU scenario from 2006 includes the early 2001 RES-E Directive measures, but still only expects increases in the share of RE of about one percentage point every five years – hitting about ten per cent in 2010 and just over twelve per cent in 2020 (European Commission, 2007a, p. 7). The second (2011) BAU scenario includes measures from the 2009 RE Directive that aim to increase the share of RE in the EU to 20 per cent by 2020 (see section 4.3.2 below). This BAU scenario outlines the achievement of the 20 per cent goal to 2020, but does not expect increases in the share of RE beyond 2020 of much more than one percentage point per five years – reaching just over 25 per cent by 2050 (Dupont & Oberthür, 2012; European Commission, 2011g).54

52 In the wake of the Fukushima nuclear disaster in Japan in March 2011, several EU member states decided to reduce or completely phase out domestic production of nuclear energy (Germany and Belgium, for example).

53 The linear trajectory for CPI may seem a simplistic tool for measuring the share of RE in final energy consumption to 2050, as early action requires high upfront costs. However, it can also be argued that early action is required to ensure GHG emissions peak early enough to mitigate climate change (IPCC, 2007). In this case, I use a linear trajectory to balance the effort evenly over time.

54 The 2011 BAU scenario here represents an average of the five BAU scenarios presented by the Commission in the 2011 impact assessment to the energy roadmap. These scenarios vary according to price of imported energy and GDP. The lowest expected share of RE in 2050 is 23 per cent – expected in a BAU scenario with low energy import prices. The highest expected share of RE in 2050 is 27 per cent – from the BAU scenario with high-energy import price assumptions (European Commission, 2011g).

108 CPI into EU renewable energy policy

Figure 4.2: Very high or complete CPI from 2000-2010, compared to BAU scenarios from 2006 and from 2011 (measured as a percentage share of final energy consumption).

Source: European Commission, 2007a; 2011g; own calculations. In the discussion on the 2001 RES-E and 2009 RE Directives below, I will be able to assess the distance between the policy output of these policies, and the very high/complete trajectory for CPI of achieving 80-100 per cent share of RE in EU energy consumption by 2050 (outlined above in Figure 4.2).

4.3 EU RE policy As described above (see section 4.1.1), the ALTENER programme was the first piece of RE policy agreed at the EU level. It was an instrument that aimed to promote mostly research and development support for the development of RE technology, but it was dramatically weakened in this mission, with proposed funds cut by the member states. The mid- to late-1990s saw much discussion in advance of the policy proposals that came in the 2000s. One of the most important communications on renewable energy was published in 1997 – the Commission’s white paper on “energy for the future” (European Commission, 1997b; Howes, 2010, p. 117).

The white paper followed on from ALTENER in describing the many co-benefits of promoting RE. These included the following points:

- RE sources can reduce energy dependency and increase energy security; - The development of RE sources can promote job creation;

109 CPI into EU renewable energy policy

- RE sources can help the EU comply with environmental protection requirements and commitments at both EU and international-level; - The growing energy consumption of developing countries provides (world- leading) EU renewable industries with an opportunity to expand (European Commission, 1997b, p. 4).

The Commission also argued in favour of policy development at the EU level to ensure an EU-wide strategy was compatible with the internal energy market. Policy at the EU level was essential “to avoid imbalances between member states or distortion of energy markets” (European Commission, 1997b, pp. 6-7).

The main objectives of the 1997 white paper included a target of doubling the share of RE in energy consumption in the EU by 2010. This meant achieving a twelve per cent share of RE by 2010 (compared to the six per cent share of RE in the early 1990s).55 This target was endorsed by the Council in May 1997, and by the European Council in June 1998 (Council of the European Union, 1997; European Commission, 2000e).56

The issue of targets was one of the first issues of contention in the ensuing discussions – and especially, whether national targets to meet the EU twelve per cent target should be binding or not. Additionally, two other issues of contention emerged in the policy discussions after the publication of the white paper: the definition of a renewable source of energy, and the proposed harmonisation of support measures for RE development (Rowlands, 2005, p. 966). The definition of what constituted renewable energy was important for whether certain types of RE would qualify for support or not (e.g. was large-scale hydropower an RE that could qualify for support?).

Of these issues, the harmonisation discussion was particularly contentious. The Commission was generally in favour of a harmonised support system for the EU – in the shape of tradable certificates of RE – rather than the most successful of the

55 In the 1990s and early 2000s, the share of RE was measured as a percentage of gross inland energy consumption, which is the total energy demand of the EU (including consumption by the energy sector itself and distribution and transportation losses). Later, the percentage share would be measured against final energy consumption, which does not include consumption by the energy sector and energy losses through transportation and distribution. The percentage share of RE in final energy consumption would be slightly greater, as it is an energy provider that does not consume fuel (see Eurostat glossary: epp.eurostat.ec.europa.eu; Howes, 2010).

56 The European Parliament rather called for a 15 per cent goal by 2010 (Jones, 2010, p. 233).

110 CPI into EU renewable energy policy national-level support mechanisms, feed-in tariffs (Boasson & Wettestad, 2013). Feed-in tariffs (or fixed-price schemes) encourage the development of RE by giving payments to RE producers for the electricity they produce (European Commission, 2000e, p. 5). Such national support schemes were in place in Germany, Denmark and in Spain, and were proving successful in encouraging the development of RE (Boasson & Wettestad, 2013, p. 82). The debate about the potential harmonisation of support schemes led to a delay in the Commission’s proposal for the eventual RES-E Directive (from 1998, as planned in the white paper, to 2000).

The Commission eventually put aside its insistence on introducing harmonised support schemes (Jansen & Uyterlinde, 2004, p. 97).57 These years between the publication of the white paper and the policy proposal for the RES-E Directive were informative for the Commission in drafting the proposal. The Commission followed the general direction of the white paper, yet adjusted its stance on harmonisation to enable agreement. The white paper was the important beginning of a debate on concrete EU-level policy measures to promote the development of RE.

4.3.1 Directive 2001/77/EC on the promotion of renewable energy sources of electricity (RES-E Directive) In the light of the heated debates and discussions that took place following the publication of the Commission’s white paper on the future of energy (European Commission, 1997b), the proposal for a directive on RE was delayed from 1998 and finally published in 2000 (European Commission, 1997b, p. 34; 2000e). The proposal came after the EU’s commitment to reduce its GHG emissions by 8 per cent by 2008- 2012 compared to 1990 levels under the Kyoto Protocol (1997) had been agreed. Policy to increase the share of RE in the EU was one of the concrete responses from the EU to achieve this target (ibid.; Boasson & Wettestad, 2013).

The proposal for a Directive on the promotion of electricity produced from renewable energy sources was published on 10 May 2000. At this point in time, RE policy focused on the electricity sector, but did not look at overall energy consumption – the heating and cooling sector was not a part of the Commission’s proposal (the transport

57 The proposal for the RES-E Directive states: “the Commission had concluded that insufficient evidence exists to provide, at this stage, for the introduction of a harmonised Community wide support-scheme” (European Commission, 2000e, p. 2).

111 CPI into EU renewable energy policy sector was dealt with in the 2003 biofuels Directive58). The legal basis in the proposal was Article 95 of the Treaty establishing the European Community (Amsterdam Treaty, TEC), placing the RES-E Directive under the internal market competence of the EU.59 This originally proposed legal competence reflects the debate within the Commission at the time, especially on the harmonisation of support schemes, and the focus on market integration. As the internal market for energy was developing (after the adoption of the 1996 electricity Directive 96/92/EC and the 1998 gas Directive 98/30/EC), the Commission feared that many different national support schemes for RE would represent barriers to fair competition and distort trade (Boasson & Wettestad, 2013, p. 84; Lauber, 2001, p. 299; Nilsson, 2011, p. 115). There was much debate on what system was most effective for promoting RE, with criticism of the Commission’s market-oriented approach coming from member states (citing the subsidiarity principle), environmental NGOs and RE industry (arguing that the quota system of the certificates would be ineffective). In the end, the Commission proposed to “monitor” the national support schemes, but did not propose a harmonised scheme (European Commission, 2000e, p. 20).

The Commission’s proposed definition of “renewable energy sources” was “renewable non-fossil sources (wind, solar, geothermal, wave, tidal, hydroelectric installations with a capacity below 10 MW, and biomass, which means products from agriculture and forestry, vegetable waste from agriculture, forestry and from the food production industry, untreated wood waste and cork waste)” (European Commission, 2000e, p. 19). The overall twelve per cent target translates into a 22.1 per cent target for RE sources in electricity production by 2010 (ibid., p 16).60 The Commission did not propose to make these targets legally binding, however, and instead requested five-yearly updates from member states on their plans for meeting the targets, with the Commission scrutinising member state progress on an annual basis. In the event that the Commission found a lack of progress in member states, it should be able to

58 Directive 2003/30/EC (the biofuels Directive) aimed to increase the share of biofuels in transport. I do not discuss this directive here, as it dealt specifically with the transport sector. The 2009 RE Directive (see section 4.3.2), however, argued for an increase in the share of RE in final energy consumption, and therefore amends both the 2001 RES-E Directive and the biofuels Directive.

59 See Annex II for the text of Article 95, TEC.

60 With the accession of new member states to the EU in 2004 and 2007, this target was revised to 21 per cent of electricity to come from RE sources by 2010 in the EU27.

112 CPI into EU renewable energy policy present proposals for imposing mandatory targets (ibid., pp. 19-20). The proposed indicative targets for the share of RE in electricity production per member state were outlined in the annex to the proposal. The biggest increase was expected from Denmark (with a target of 29 per cent of RES-E in 2010, to be increased from an 8.7 per cent share of RE in electricity in 1997).

In its proposal, the Commission describes that discussions took place with member state agencies, industry, professional associations and NGOs, as well as internal discussions among the Commission services in advance of the publication of the proposal (European Commission, 2000e, p. 13). In all of these consultations, the focus was on support schemes. Consultation also took place with the electricity supply industry and associations representing the RE industry on issues of administrative barriers and grid reinforcement requirements for integrating more RE into the electricity system (ibid.).

The Parliament’s energy committee (ITRE) was in charge of drafting the first reading response to the RES-E proposal. MEP Mechtild Rothe of the European Socialists political group was the rapporteur. MEP Hans Kronberger (no political group) provided an opinion on behalf of the ENVI committee. The first reading report was adopted by the Parliament on 16 November 2000.

The Parliament suggested 68 amendments to the Commission’s proposal (European Parliament, 2000b). The main substantive changes included the Parliament’s disagreement with the Commission’s calculations. The Parliament argued that the 12 per cent overall share by 2010 translated to a 23.5 per cent share of RE in electricity generation (as also mentioned in the 1997 white paper; European Commission, 1997b), and requested a change in targets in this respect. Additionally, the Parliament called for the national targets to be mandatory. Finally, the Parliament supported national-level support schemes. The rapporteur explained that there was a positive experience with the support schemes, and that they were necessary because of historical support to conventional and nuclear energy production that led to unfair competition for the new RE producers (ibid.).61 Generally, the Parliament aimed to

61 A proposal from the ENVI committee to include peat as part of the definition of renewable energy sources is surprising (given that peat is a GHG emitting solid fuel), but this featured in the Parliament’s final report nonetheless (European Parliament, 2000b, amendment 27). The Commission did not accept this particular amendment (European Commission, 2000b, p. 2).

113 CPI into EU renewable energy policy strengthen the proposal and ensure trade and competition concerns did not hamper the increase in the share of RE.

The Commission responded with an amended proposal in December 2000 (European Commission, 2000b). It accepted a number of the Parliament’s proposed amendments, including a definition for biomass to include a proviso that biomass “with insignificant impurities” would be considered renewable (ibid., p. 9). The Commission did not accept Parliament’s calculation of 23.5 per cent share in electricity (ibid., p. 5), nor did it take on board amendments making the national targets mandatory, but kept the wording as follows: “…require all member states to set national targets…” (ibid., p. 6). The Commission otherwise accepted much of the wording of the Parliament about evaluating national support schemes.

The next step in the legislative proposal included the Council’s common position, which was published on 23 March 2001 (Council of the European Union, 2001). The Council placed the proposal under the environmental chapter of the Treaty (Article 175(1), TEC),62 rather than under the internal market, as proposed by the Commission. The Council clarified that the targets should be non-binding, stating “all member states should be required to set national indicative targets for the consumption of electricity produced from renewable sources” (Council of the European Union, 2001, p. 3). When drawing up their own “indicative” targets, member states had to “take account of” the national targets drawn up per member state by the Commission in the annex to the proposal (ibid., p. 9). Additionally, the Council removed the limit on hydropower in the definition of renewable energy sources (from a maximum of 10 mega watts), so that even large-scale hydropower could be considered a renewable energy source. The Council also removed Parliament’s amendment that only biomass “with insignificant impurities” (European Parliament, 2000b) could be considered renewable.

The proposal moved into second reading in the Parliament, with the ITRE committee’s second reading recommendation tabled in the plenary on 20 June 2001 (European Parliament, 2001a). Media predicted a tough battle between Parliament and Council in the second reading, with MEP and rapporteur Mechtild Rothe warning

62 See Annex II for the text of Article 175(1), TEC.

114 CPI into EU renewable energy policy that she would stick to the proposal for legally binding targets (ENDS Europe, 2001c). Informal negotiations between the then Swedish Presidency and the rapporteur resulted in a deal aimed to avoid entering the conciliation procedure (ENDS Europe, 2001a). The negotiated deal meant that the non-binding targets remained in exchange for a commitment that binding targets would be introduced in future if the indicative approach fails. As amended by the Parliament in its second reading, recital seven of the final Directive reads: “… if necessary for the achievement of the targets, the Commission should submit proposals to the European Parliament and the Council which may include mandatory targets”. Agreement was also finally reached between Parliament and Council to allow energy produced from waste incineration to be defined as a renewable energy, and hydropower (without any upper capacity limit) was also included (Art. 2). The final act was signed by Council and Parliament on 27 September 2001.

4.3.1.1 CPI in the policy process of the 2001 RES-E Directive As discussed in chapter three (see Table 3.1), measuring the level of CPI in the policy process involves examining three main factors: the involvement of internal pro- climate stakeholders in the EU institutions; the involvement of external pro-climate stakeholders in the policy process; and the recognition of the functional interrelations between the policy objectives of (in this case) RE policy and the long-term climate objectives to reduce GHG emissions in the EU by 80 to 95 per cent by 2050 compared to 1990 levels (see also section 4.2.1). For each of these factors, a qualitative assessment, based on the indicators outlined in Table 3.1, will lead to an overall level of CPI in the policy process leading to the 2001 RES-E Directive. Each of these three factors is discussed separately below.

Internal pro-climate stakeholders In the Commission, DG Transport and Energy (DG TREN: today (and hereafter): “DG Energy”) drafted the proposal for the RES-E Directive. This proposal was published after much discussion within the Commission, and between the Commission and member states, following the publication of the 1997 white paper on the future of energy (European Commission, 1997b; 2000c, p. 13; see above). Much of the discussion within the Commission itself circled around the idea of harmonisation of support schemes, which led to delays in the drafting of the proposal (ENDS Europe, 1999c). In the end, with member states, RE industry associations,

115 CPI into EU renewable energy policy environmental NGOs and MEPs all arguing against harmonisation of support schemes, the Commission accepted to “monitor” national support schemes (European Commission, 2000c; Boasson & Wettestad, 2013). There is no clear evidence of the pro-active involvement of DG Environment in the development of the Commission’s proposal, but they were certainly involved in the long discussions before the proposal was published (ibid.).

In some respects, DG Energy itself can be said to have acted as a somewhat pro- climate stakeholder in the policy process. There may not have been any further need for the involvement of DG Environment, as the message from the Commission was for advancing RE policy in order to achieve climate objectives, at least in the short- term (under the Kyoto Protocol). In the proposal, the Commission states: “In view of the substantial contribution RES can make to the implementation of the Community’s commitments to reduce greenhouse gases [sic], its expansion in the EU constitutes an essential part of the package of measures needed to comply with the Kyoto Protocol…” (European Commission, 2000e, p. 3). As suggested in Table 3.1, this implies a medium level of CPI in the policy process for the Commission. Although DG Environment were not “co-drafters” of the proposal, they were involved in discussions during the proposal’s development, and in the end, DG Energy took on the role of an internal pro-climate stakeholder in advancing RE policy for the sake of achieving climate policy objectives.

In the Parliament, although the ITRE committee drafted the report, the ENVI committee provided an opinion for the rapporteur. In this case, some of the ENVI suggestions were taken on board, and overall, it can be said that Parliament was ambitious in pushing for a policy measure that would make a difference, at least in the first reading. The Parliament pushed for mandatory national targets and highlighted the environmental and climate policy objectives of strong RE policy. The rapporteur’s first reading report explained the many advantages of pushing for ambitious RE policy as follows:

“Renewable energies are an integral feature of an effective strategy to protect the climate; they help achieve the Kyoto objectives; they do not waste resources; they reduce emissions of harmful substances into the air; they make it possible to develop a decentralised structure of energy supplies, together

116 CPI into EU renewable energy policy

with the possibility of sustainable regional development and new employment prospects; they create security of supplies; and, in an international context, they help developing countries solve a variety of problems” (European Parliament, 2000b, p. 38).

Here, it is clear that the climate and environmental objectives that can be achieved through increasing RE are of paramount importance for the ITRE committee, with co- benefits for job creation and energy security coming later in the list of benefits. It could thus be said that the rapporteur in this case (MEP Mechtild Rothe) was a pro- climate stakeholder herself, although working through the ITRE committee. The ENVI committee did have the opportunity to provide an opinion and influence the final ITRE committee report. ENVI’s amendments highlighted the role of the RES-E Directive in meeting GHG emission reduction targets under the Kyoto Protocol, which was accepted in the final legislative act. Other proposals, taken on board also by the ITRE committee (such as ensuring binding national targets) did not survive the policy negotiations. In this case, at least for the first reading, the ENVI committee provided several points that were taken on board by the ITRE committee in the drafting of the first reading report. However, negotiations on the second reading agreement took place informally between the rapporteur and the Presidency (ENDS Europe, 2001a). The ENVI committee had no role in these informal negotiations. Even though MEP Mechtild Rothe aimed to keep her strong stance throughout the second reading (ENDS Europe, 2001c), she finally relented and accepted many member state demands for the sake of agreement (see also section 4.3.1). In accordance with Table 3.1, the involvement of internal pro-climate stakeholders in the Parliament is medium. This is because, although ENVI provided an opinion in the first reading report, and several of ENVI’s proposed amendments were included, ENVI did not participate in the informal negotiations that followed in second reading. While the rapporteur originally claimed she would stick to the objective of making the targets mandatory, she finally compromised with member states in exchange for promises that proposals for mandatory targets could be made in future if the indicative approach failed.

In the Council, negotiations took place within the energy Council formation. There is no apparent interaction with environment ministers (although it is true that negotiations in Council are often behind closed doors). Even though the Council and

117 CPI into EU renewable energy policy the European Council supported the objective to increase the share of RE in the EU to twelve per cent by 2010, as outlined in the 1997 white paper (see above), member states were not prepared to agree to mandatory targets. This, and the issue of the definition of sources of renewable energy, seemed to be the main sticking points in the negotiations with Parliament (ENDS Europe, 2000d). Although there were hopes from the French Presidency in the second half of 2000 to reach rapid agreement within their term (ENDS Europe, 2000b), these issues delayed agreement until 2001. In early 2000, some member states already raised questions about their shares of the indicative targets, calling for them to be reduced (ibid.). Italy pushed for waste incineration to be included as part of the definition of renewably-sourced electricity – a proposal that Parliament would not accept in early stages of the negotiations (ENDS Europe, 2000d). Finally, the Council did push for waste incineration as part of the definition and supported indicative targets. Portugal, Finland and the Netherlands negotiated for lower indicative targets to 2010 as the proposed targets were considered too ambitious (Council of the European Union, 2001; ENDS Europe, 2000a). Unlike in the Parliament and Commission, the absence of pro-climate stakeholders in the policy process may have lowered the level of CPI overall in the policy process in the Council. In this case, (and in line with Table 3.1) there is no or very low involvement of internal pro-climate stakeholders in the policy process.

Taking an aggregate of the medium levels of involvement of pro-climate stakeholders in the Commission and Parliament with the very low levels in the Council, the overall level of involvement of internal pro-climate stakeholders in the process leading to the agreement on the RES-E Directive is low to medium.

External pro-climate stakeholders The Commission’s proposal for the RES-E Directive was published in 2000, before requirements for an impact assessment and open public consultations were required.63 There was no impact assessment procedure and no open public consultation leading to the publication of the Commission’s proposal. Nevertheless, there was much discussion among EU institutions, member states and other stakeholders in the aftermath of the publication of the 1997 white paper. The Commission reports that

63 These became a regular part of the Commission’s preparations for policy proposals after 2002, when better regulation and minimum standards for consultation were adopted by the Commission (European Commission, 2002d).

118 CPI into EU renewable energy policy industries, professional associations and NGOs were involved, along with Commission officials and member state representatives, in the discussions on this paper (European Commission, 2000e, p. 13). RE industry representatives were consulted, especially with regard to issues such as administrative procedures and grid issues (ibid.).

Without formal consultation procedures in place, it is difficult to assess the extent to which external pro-climate stakeholders had access to the Commission. One interviewee suggested that in the late 1990s and early 2000s, environmental NGOs, in particular, were less a part of the policy process in general than in the mid- to late- 2000s (interview 7), but that even later, the Commission was not the first lobbying target for environmental NGOs. The RE industry at this time was focused on ensuring the national support schemes remained in place, and rather worked with the Parliament and through their national governments than with the Commission (Boasson & Wettestad, 2013; Hildingsson, et al., 2010). The German RE industry was particularly strong on pushing the German government not to accept Commission proposals for harmonising support schemes (ENDS Europe, 1999c). A coalition of environmental groups did try to influence the Commission’s proposal by calling jointly for a strong RE law and pushing EU member states to agree to a 16 per cent RE share target for 2010 (ENDS Europe, 1999d). The coalition, including Greenpeace and WWF, rather targeted member states than the Commission directly. Later, the NGOs WWF and Climate Action Network criticised the Commission’s low ambition in its proposals (ENDS Europe, 1999a). It seems that the informal lobbying of environmental NGOs in the advance publication of the proposal did little (if anything) to change the policy direction towards a more ambitious level.

Thus, in line with Table 3.1, the involvement of external pro-climate stakeholders in the policy process with the Commission is low. There are no official procedures in place to allow external stakeholders access to the Commission, and consultation seems to have taken place on an ad hoc basis. The involvement of environmental NGOs is unclear, and the RE industry preferred to lobby their own governments first.

In the Parliament, environmental NGOs generally agree that they have easy access, and better chances to make their voices heard (interviews 5, 6, 7, 8 & 11). In the early 2000s, environmental NGOs backed up the Parliament’s first reading stance on the

119 CPI into EU renewable energy policy

RES-E proposal: namely, to push for mandatory targets and to ensure that national support schemes continued (as national support schemes were seen as more favourable to advance RE). At this time, the RE industry was becoming increasingly organised at the EU level (in the form of the European Renewable Energy Council, EREC; Boasson & Wettestad, 2013). The Parliament was generally an ally of environmental NGOs and the RE industry, who were reportedly happy with the result of the first reading (ENDS Europe, 2000c). The same external pro-climate stakeholders were, however, generally dissatisfied with the final result of the policy negotiations (Nilsson, 2011). Nevertheless, although Parliamentary actors were overall in favour of pushing for more ambitious policy on RE, there were no official procedures for consultation. All involvement of external pro-climate stakeholders was on an informal basis. While interviewees considered their access to the Parliament as sufficient (interviews 1, 5, 6, 7, 8, & 11), their access to policymaking processes became more limited as the process continued. Informal negotiations in the second reading effectively excluded external pro-climate stakeholders. They could not access the policymaking process and neither could they follow the developments in the process, as these negotiations took place behind closed doors (interview 6).

According to Table 3.1 this situation amounts to a low to medium level of involvement of external pro-climate stakeholders in the policy process with the Parliament. There were no official procedures in place, but all interviewees reported easy access to the Parliament. However, this ease of access was only in the early stages of the policy process, as informal negotiations between Council and Parliament in the second reading excluded external stakeholder involvement.

In the Council, there were very few opportunities for external stakeholders to get involved in the policy process. This is due to the fact that many negotiations in the Council take place behind closed doors, and are not transparent (interview 6). Environmental NGOs were particularly limited in this case in their ability to access the Council. This often requires lobbying governments at the member state level, rather than at the EU level. Several interviewees commented that their own internal resources did not always permit environmental NGOs to follow such a strategy (interviews 5, 6 & 7). The RE industry, however, lobbied at both member state and EU level (see above). RE industries were organised at EU level under EREC, but they did not benefit from any special access to the Council. The push for binding RE

120 CPI into EU renewable energy policy targets, for example, did not survive in the Council (Council of the European Union, 2001).

The involvement of external pro-climate stakeholders in the process with the Council can only be considered low. While in the early stages, RE industries in some member states could access their national governments and push policy at that level, in later stages of the policy negotiations, there was no access to the Council for external environmental NGOs or RE industry representatives. These stakeholders tried to remain visible throughout the process, but could not access a closed Council.

In summary, the involvement of external pro-climate stakeholders in the policy process leading to the adoption of the 2001 RES-E Directive is generally low, although informal access to the Parliament scored low to medium (see Table 4.2).

Recognition of functional interrelations From the very beginning of policy discussions on the development of RE policy in the EU, policymakers highlighted the benefits for climate policy of agreeing measures to increase the share of RE in the EU. The Commission stated in its proposal that one of the main objectives of the RES-E Directive was to meet “the obligation to reduce the emission of greenhouse gases accepted by the EU at Kyoto” (European Commission, 2000e, p. 2). The Parliament also highlighted, especially, the climate and environmental objectives of increasing the share of RE in the EU, and aimed to strengthen the Commission’s proposal (European Parliament, 2000a). The Council and European Council had originally supported the Commission’s 1997 white paper on renewable energy, which called for a doubling of the share of RE in the EU to twelve per cent by 2010 as a response to climate commitments (European Commission, 1997b). Yet in the ensuing negotiations on the RES-E Directive, member states watered-down the proposal to ensure targets remained “indicative” and that waste incineration would be defined as a renewable source of energy (Council of the European Union, 2001).64

64 The Council did propose a change in the legal basis for the RES-E Directive from Article 95 (internal market) to Article 175 (environment) of the TEC. RE policy is then firmly located in EU legal competence on improving the state of the environment and combating climate change. However, this could be seen rather as a strategy to move the policy out of internal market concerns and make it more difficult in future for the Commission to propose harmonisation of support schemes (Council of the European Union, 2001).

121 CPI into EU renewable energy policy

The synergetic functional interrelations between RE policy objectives (to increase the share of RE in EU energy consumption) and climate policy objectives (to reduce GHG emissions) were certainly recognised by policymakers in the early stages of the policy process especially. These functional interrelations were rather limited to short- term climate policy objectives (such as the commitments under the Kyoto Protocol for 2008-2012), and there is little mention of long-term climate policy objectives to limit global temperature increase to two degrees Celsius. The final Directive does not provide for mandatory targets, and neither can the output (twelve per cent RE share by 2010) be considered ambitious enough (see section 4.3.1.2). Policy negotiations in the later phases pitted the Parliament (which was pushing for mandatory targets and ambitious policy) against the member states (who were in favour of indicative targets only and a broad definition of RE sources). In some respects, the functional interrelations remained a stronger part of the negotiation strategy of the Parliament, yet the rapporteur nevertheless compromised with the Council in informal negotiations in the second reading. Perhaps for the sake of getting agreement, the long-term functional interrelations (which, if recognised, should help push for ambitious RE policy), were rather left aside in the later stages of the policy negotiations.

As outlined in Table 3.1, the recognition of functional interrelations between long- term climate policy objectives and RE policy objectives in the policy process leading to the RES-E Directive is situated close to the medium level (functional interrelations recognised and considered by policymakers in the process, but this does not motivate strong policy action). This is because the co-benefits of the policy objectives were highlighted from the outset of the policy proposal, yet relating more to short- or medium-term climate policy objectives than long-term objectives. The recognition of these functional interrelations was also insufficient to push policy in an ambitious direction.

Taken together, the low to medium levels of involvement of internal pro-climate stakeholders, the low levels of involvement of external pro-climate policy stakeholders, and the medium recognition of the functional interrelations in the policy process suggests an overall level of CPI in the policy process in the 2001 RES-E Directive of low to medium.

122 CPI into EU renewable energy policy

4.3.1.2 CPI in the policy output of the 2001 RES-E Directive As shown in Figure 4.2, the very high or complete levels of CPI for RE policy would see an increase in the share of RE in EU final energy consumption by between 7 and 9 percentage points every five years from 2000 to meet long-term climate objectives by 2050. The data for Figure 4.2 refers to final energy consumption, which does not include consumption by the energy sector itself or losses from distribution and transportation. The 2001 RES-E Directive, however, outlines an indicative target to achieve a twelve per cent share of RE in gross inland energy consumption (including energy sector consumption) by 2010, which corresponds to about a 22 per cent share of renewably generated electricity by 2010. Whether describing the share of RE in final energy consumption or gross inland energy consumption, the very high levels of CPI outlined in Figure 4.2 can still apply – namely achieving an 80 to 100 per cent of RE by 2050. The EEA has reported on shares of RE in gross inland energy consumption in the EU27 from 1990.65 This data can provide suggestions of what BAU between 1990 and 2010 may have looked like without the RES-E Directive in place. Plotted against this, the twelve per cent target as the policy output for the RES- E Directive to 2010 can show the distance covered by this policy measure towards closing the gap between BAU and very high levels of CPI.

According to the EEA, the share of RE in gross inland energy consumption in the EU27 stood at 4.2 per cent in 1990, rose to five per cent in 1995 and to 5.6 per cent in 2000.66 This implies that the share of RE in the EU was on the increase before the 2001 RES-E Directive was agreed, but only by about 0.7 percentage points every five years. Assuming this trend would have continued as BAU between 2000 and 2010, there would have been an increase in RE share in gross inland energy consumption to 6.3 per cent in 2005, and to seven per cent in 2010 (European Commission, 1996, p. 15). This is clearly below the 2001 RES-E Directive’s target. With RE share of 5.6 per cent in 2000, meeting the 80 to 100 per cent CPI targets suggests a required percentage share of between 21 and 25 per cent share in RE by 2010 (see Figures 4.2 and 4.3). Thus, although the 2001 RES-E Directive’s policy output succeeds in closing the gap between BAU and CPI trajectories by five percentage points, this

65 See: http://www.eea.europa.eu/data-and-maps/indicators/renewable-primary-energy-consumption/renewable-primary-energy- consumption-assessment-7, accessed: 12 March 2013.

66 These figures are similar to the share for the EU15 (4.7 per cent in 1990; 5.1 per cent in 1995; and 5.6 per cent in 2000).

123 CPI into EU renewable energy policy amounts to closing the gap by between 28 and 36 per cent. In other words, the policy output of the 2001 RES-E Directive is between 28 and 36 per cent in line with long- term climate policy objectives to decarbonise by 2050. According to Table 3.2, this implies low levels of CPI for the policy output in the 2001 RES-E Directive (see Figure 4.3 below).

Figure 4.3: Policy output of the 2001 RES-E Directive compared to BAU and to very high levels of CPI (measured as a percentage share of gross inland energy consumption).

Source: EEA database, www.eea.europa.eu; own calculations. As the targets of the 2001 RES-E Directive were indicative, it is interesting to compare the stated policy objective against the implementation reality. The same EEA data shows that in 2009, the RE percentage share in the EU27’s gross inland energy consumption stood at nine per cent.67 According to Commission data from 2011, this nine per cent share remained in 2010, so the twelve per cent target was not met (European Commission, 2011g). In this case, the actual implementation of the 2001 RES-E Directive increased the share of RE by just two percentage points compared to BAU, implying the RES-E Directive closed the gap between BAU and complete CPI by just 11-14 per cent (very low levels of CPI, see Table 3.2).

67 Figures for percentage RE share in final energy consumption for 2010 show over 11 per cent RE in the EU27. It is clear that in terms of measuring the share of RE energy, the data tell a more positive story when we look at final energy consumption. As RE does not consume fuels like other energy producers, RE share in final energy consumption is considered a more appropriate measuring tool, and this is taken up for the next major piece of EU legislation on RE in 2009.

124 CPI into EU renewable energy policy

These very low levels of CPI in the policy output of the 2001 RES-E Directive could nonetheless also be considered too harsh a measurement. It would be useful to consider the levels of CPI in the policy output in terms of capacity and potential for achieving a speedy increase in the share of RE at the time. In its 1996 green paper on the future of energy, the Commission outlined one scenario that suggested about 13 per cent share of RE in gross inland energy consumption would be possible with strong policies by 2010. Considering that nine per cent was achieved without mandatory targets (without what could be called strong policy), a 13 per cent share seems rather low. Nevertheless, it is not unrealistic to argue that RE technology was still costly in the early 2000s. Making an assumption that a steeper climb to the 2050 targets would begin after 2010, as costs for RE fall, would provide a different scenario. If the twelve per cent share was achieved in 2010, it would imply between about nine and eleven percentage points increases every five years from 2010 to meet the 2050 target. This may be considered quite a jump from the one to two percentage point increases every five years from 1990 to 2010. Taking instead a target, as argued for by the European Parliament (see above) of 15 per cent RE share in 2010 would call for between eight and nearly eleven percentage point increases per five years beyond 2010. A mid-way point between the complete CPI figures of 21 and 25 per cent RE share by 2010, and the actual twelve per cent target would suggest between about 17 and 19 per cent RE share in 2010 (leaving between eight to ten percentage point increases required per five years from 2010 to 2050). Figure 4.4 outlines these midway ranges for 2010 targets for CPI to 2010 against the policy output of the 2001 RES-E target and against the BAU scenario. Taking this more tempered benchmark into account, the policy output of the 2001 RES-E Directive closes the gap between BAU and very high CPI levels by between 42 and 50 per cent. Such a level of CPI in the policy output is medium.

However, considering the expected implementation deficit of many pieces of EU legislation, especially when the targets are non-binding (Haverland & Romeijn, 2007; Lampinen & Uusiklyä, 1998), and considering the empirical advantage available for the RES-E Directive to assess ex ante the performance, even a generous level of medium is misleading. The final nine per cent share of RE in the EU27 in 2010 (see above) closes the gap between BAU (seven per cent in 2010) and a tempered CPI level (between 17 and 19 per cent in 2010) by between only 20 and 25 per cent. This

125 CPI into EU renewable energy policy level, according to Table 3.2 indicates low levels of CPI in the policy output of the 2001 RES-E Directive.

Figure 4.4: CPI in the policy output of the 2001 RES-E Directive, assuming lower advances in RE share between 2000 and 2010 than between 2010 and 2050.

Sources: (European Commission, 1996, 2011g); own calculations.

Overall, the level of CPI in the policy output of the 2001 RES-E Directive is low at best, considering the non-binding nature of the (already insufficient) target, the recognised implementation deficit for EU legislation, and considering the final implementation outcome of the Directive (see Table 4.2).

4.3.2 Directive 2009/28/EC on the promotion of the use of energy from renewable sources (RE Directive) The Commission reported regularly on the progress in achieving the targets set out in the 2001 RES-E Directive in the years that followed. In its 2006 review, the Commission highlighted the less-than-perfect implementation and paved the way for future policy measures. By this time, it had already begun infringement proceedings against six member states68 for reasons of incomplete transposition of the Directive into national law; lack of commitment on the targets; lack of implementation of the guarantees of origin certificate system; lack of transparency in administrative procedures to issue licenses for new renewable electricity plants; and lack of transparency regarding access to grids and regarding rules on grid investment

68 Austria, Cyprus, Greece, Ireland, Italy and Latvia (European Commission, 2006d).

126 CPI into EU renewable energy policy

(European Commission, 2006d, p. 18). As a result of the poor performance in improving the penetration of RE sources of electricity, the Commission announced it would propose a new legal framework for RE in a roadmap (European Commission, 2006d, p. 19).

The promised renewable energy roadmap was published in 2007. It aimed to establish a long-term vision for RE in the EU, and paved the way for future legislation on the issue (European Commission, 2007b). It highlighted the unlikelihood of meeting the 2010 target, which constitutes a “policy failure and a result of the inability or the unwillingness to back political declarations by political and economic incentives” (European Commission, 2007b, p. 8). Most importantly, the roadmap contained a proposal for a legally binding target of 20 per cent share of RE in the EU’s energy consumption by 2020 – a target that remained throughout the negotiation process and became part of the final text of Directive 2009/28/EC (European Commission, 2007b, p. 3). This target came in response to scenarios and assessments in light of the European Council’s call for a target of 15 per cent share of RE by 2015, and the European Parliament’s call for a 25 per cent target by 2020 (European Council, 2006, p. 15; European Parliament, 2006b). The roadmap also outlined a departure from earlier legislation by highlighting the need for the future policy framework to cover RE sources in electricity, transport and also the heating and cooling sector under one policy instrument (with the RE target then being measured as a proportion of final energy consumption).

The Commission put forward its proposal for a Directive on the promotion of the use of energy from renewable sources in January 2008. It was proposed as part of the “climate and energy package” (European Commission, 2008d) that also included a proposal for revising the emissions trading system (ETS), a proposal for a decision on reducing GHG emissions in sectors not covered by the ETS, and a proposal for a Directive to support carbon capture and storage (CCS) technology (Oberthür & Pallemaerts, 2010a, 2010b). The proposal aimed to establish the 20 per cent RE target for 2020 as a binding target, and a second target to increase the share of biofuels and renewables in transport to ten per cent by 2020 (European Commission, 2008d, p. 8). It was proposed on the dual legal basis of Article 175(1) TEC of the environment

127 CPI into EU renewable energy policy chapter and Article 95 TEC on the internal market.69 The Commission justifies this dual legal basis as the proposed articles on biofuels and bioliquids prevent member states from adopting measures that would block trade in biofuels. The rest of the proposed Directive is considered to fall under the objectives of the environment chapter to “preserve, protect and improve the quality of the environment, protect human health and make prudent and rational use of natural resources” (European Commission, 2008d, p. 8).

The consultations leading up to the proposal included public consultations on the 2007 RE roadmap, on the energy green paper (European Commission, 2006b), and on the strategic energy review (European Commission, 2007c) between March and September 2006. In addition, public consultations took place in 2007 with “member states, citizens, stakeholder groups, civil society organisations, NGOs and consumer organisations” (European Commission, 2008d, p. 5). The major issues touched upon in these public consultations included a review of the promotion of biofuels in new RE legislation; the promotion of heating and cooling from RE; and administrative barriers to the development of RE. The Commission reported general support for stronger policy and long-term goals in RE policy, and also support for sustainability criteria for promoting biofuels (European Commission, 2008d, pp. 5-6).

The public consultation on biofuels included 62 responses from NGOs, of which just over a third came from environmental NGOs. These included responses from environmental NGOs outside the EU also, including a coalition of five African environmental NGOs70 and Friends of the Earth, Malaysia. The other NGO responses came from biofuels consultancy organisations, associations of industries either producing or consuming biofuels, or associations concerned with food production displacement of a potential increase in biofuels production.71 The consultation on heating and cooling included just two responses from NGOs, but 20 from industry (out of a total of 117 responses; Mercier & Peteves, 2006, p. 3). In the consultation on

69 See Annex II for the text of Articles 175(1) and 95, TEC. As noted above in section 4.3.1, the Commission originally proposed the 2001 RES-E Directive under Article 95, but it was later adopted with Article 175(1) as its legal basis.

70 A joint response from the Africa Biodiversity Network, Kenya; Melca Mahiber, Ethiopia; Envirocare, Tanzania; Climate and Development Initiatives, Uganda; Nature Tropicale, Benin.

71 Over 70 responses came from public institutions and member states. A further 57 responses came directly from industry actors. Another 54 responses were from citizens, and the Malaysian government provided its own response. See: http://ec.europa.eu/energy/renewables/consultations/2007_06_18_biofuels_en.htm, accessed: 10 February 2012.

128 CPI into EU renewable energy policy the energy green paper (in advance of the second strategic energy review), a stable policy framework with a mandatory target for RE was supported by the majority (European Commission, 2006f, pp. 3-4). This consultation received 1,516 responses to the online questionnaire and 164 additional written comments, 22 of which came from NGOs and 91 from industrial associations and companies (European Commission, 2006f, p. 8).

The process leading to the adoption of the 2009 RE Directive took place in one reading. The negotiations among the co-deciding institutions began informally early after the publication of the Commission’s proposal. In the Parliament, MEP Claude Turmes (Greens) was the rapporteur for the ITRE committee on the dossier, but the ENVI committee had a special role as an associated committee for the first reading report (meaning it was jointly responsible with the ITRE committee, and did not simply provide an opinion).72 Anders Wijkman of the EPP group was the rapporteur for the ENVI committee. MEPs highlighted three main issues in the proposal that they would like to see strengthened: the interim targets for member states to meet the 2020 target (including sanctions if such interim targets were not met); sustainability criteria for biofuels in transport, if that target was to remain; and priority access for renewables to the electricity grid (ENDS Europe, 2008m). A fourth issue, that was a sticking point also for Council, was on the renewables trading certificates (ENDS Europe, 2008b, 2008h). While the rapporteur considered the proposed Guarantees of Origin (GO) certificate trading too legally ambiguous, member states (also supported by the Parliament) called instead for more flexibility on renewables trading (ENDS Europe, 2008b; Nilsson, Nilsson & Ericsson, 2009).

Throughout 2008, negotiations between Parliament and Council continued in trilogues. The proposal was discussed in the energy Council formation on 28 February, 6 and 9 June and 8 December 2008. Although the environment Council discussed the climate and energy package on 3 March, 5 June, 20 October and 4 December 2008, it focused on the ETS, effort-sharing decision and CCS Directive proposals. These were the three legislative proposals considered within the “competence” of environment ministers (Council of the European Union, 2008c, p. 9). The main points of

72 The agriculture and rural development; the regional development; the international trade; the transport and tourism; and the economic and monetary affairs committees all provided opinions on this file.

129 CPI into EU renewable energy policy contention for member states in the negotiations on the RE Directive included the strength of the national targets (and the consequences of not meeting the indicative trajectory), the sustainability criteria for the biofuels target and the system of GO trading (Council of the European Union, 2008a, 2008e; ENDS Europe, 2008b).

In negotiations in September 2008, the ITRE committee insisted on mandatory interim targets for member states, with penalties for missing these targets (ENDS Europe, 2008i). The committee was ready to accept Council’s proposal for more flexibility in renewables trading, and also to accept the ten per cent target for biofuels and renewables in transport. However, the committee included certain criteria for biofuels before they could be considered to meet the target (i.e. at least two-fifths of the overall share should be from second-generation biofuels or electric vehicles powered from RE; ibid.). A further review clause on the biofuels provisions for 2014 was included. The rapporteur praised the inclusion of this review clause as an opportunity to strengthen the Directive later (ENDS Europe, 2008j).

By the end of October 2008, member states had reached an agreement within the Council, and negotiations on the remaining sticking points with Parliament continued. The major remaining issue at this point seemed to be about the sustainability criteria for biofuels (Council of the European Union, 2008b; ENDS Europe, 2008f). Additionally, a review of all the targets in the Directive for 2014 was pushed for by Council – some member states saw this review clause as an opportunity to reconsider the strength of the targets overall (Council of the European Union, 2008e). Compromise between the Parliament and Council was reached by early December. Interim targets to 2020 remained indicative and no automatic financial sanctions were included for member states that miss their binding 2020 target. Agreement was reached that measures on the indirect land use changes due to biofuel production would be adopted later and the Commission was granted oversight on the member state renewable action plans to be submitted after the Directive was adopted. Despite Italian reticence, the 2014 review clause clarifies that no changes can be made to the RE targets as a result of the review (ENDS Europe, 2008e, 2008l).

With the RE Directive forming part of the climate and energy package, it was negotiated alongside the revised ETS Directive (2009/29/EC), the Directive on CCS (2009/31/EC) and the effort-sharing Decision for emissions not covered by the ETS

130 CPI into EU renewable energy policy

(Decision 406/2009/EC). On 12 December, in an unusual move under the co-decision procedure, the European Council announced agreement on the entire package of legislative measures (ENDS Europe, 2008g; European Council, 2008b, pp. 8-9). This announcement came before the Parliament adopted its first reading agreement, on 17 December 2008 (although the positions were already aligned, the Parliament had not yet officially adopted its position in plenary when the European Council announced agreement). The Commission accepted the text and the Directive was signed into law on 24 April 2009 as Directive 2009/28/EC.

4.3.2.1 CPI in the policy process of the 2009 RE Directive As above, and as outlined in Table 3.1, I here discuss three indicators to measure the level of CPI in the policy process: the involvement of internal pro-climate stakeholders in the EU institutions; the involvement of external pro-climate stakeholders in the policy process; and the recognition by policymakers of the functional interrelations between the policy objectives of (in this case) RE policy and the long-term climate policy objectives to reduce GHG emissions in the EU by 80 to 95 per cent by 2050, compared to 1990 levels.

Internal pro-climate stakeholders As the RE Directive was negotiated under the co-decision procedure, consultation procedures were in place (this means internally within the EU institutions and with external stakeholders through open, public consultations). In the Commission, DG Energy drafted the proposal on the RE Directive. Usual inter-service consultation procedures applied allowing space for DG Environment (and all other DGs) to lend its voice to the drafting. Additionally, the RE Directive was proposed as one part of the climate and energy package of policy measures in early 2008. Other measures in this package (such as the proposal for a Directive on CCS, and proposal for a revised ETS Directive) were under the responsibility of DG Environment. There was enhanced cooperation between DGs Energy and Environment on the preparation of these draft proposals to publish a coherent package of policy measures in early 2008 (Boasson & Wettestad, 2013, p. 45). As such, the level of involvement of internal pro-climate stakeholders in the Commission for the 2009 RE Directive is (according to Table 3.1) medium-high. This is because DGs Energy and Environment worked closely together in the drafting of the proposals in the climate and energy package (thus a high involvement), but DG Energy was solely responsible for the RE Directive proposal

131 CPI into EU renewable energy policy

(thus a medium involvement of internal pro-climate stakeholders throughout the process).

In the Parliament, the ITRE committee was responsible for drafting the first reading report on the proposal, but the ENVI committee was associated with the ITRE committee for the drafting procedure. Although ITRE was still the leading committee, the associated committee status for ENVI implies high collaboration between the committees on this particular dossier.73 The rapporteurs of the ITRE committee (MEP Claude Turmes) and the ENVI committee (MEP Anders Wijkman) worked closely together under the associated committee procedure to finalise a first reading report. Bolstering the internal pro-climate stakeholder involvement in the Parliament further, the rapporteur on this dossier was a member of the Greens political group. MEP Claude Turmes pushed amendments that strengthened the sustainability criteria for meeting the ten per cent target in transport, and reached a compromise solution with the Council on the review of the Directive scheduled in 2014, that it would not involve a review to lower the ambition of the targets. With the ENVI committee holding the associated committee status, and with the ITRE committee’s rapporteur being a member of the Greens political group, the involvement of internal pro-climate stakeholders in the Parliament is high (see Table 3.1).

In the Council, the link between environment and energy Council formations is due to the proposal for the RE Directive being made as part of the climate and energy package. The package was discussed in both the energy and environment Council meetings throughout 2008, but there was a clear division of labour in the Council, with the energy Council formation being responsible for the RE Directive. It was the energy Council under the French Presidency in the second half of 2008 that negotiated informally with Parliament (with rapporteur Claude Turmes; Boasson & Wettestad, 2013; Council of the European Union, 2008d; ENDS Europe, 2008e). Although the overall progress of the climate and energy package was reported to both the energy and environment Council formations, the substantive discussions on the RE Directive remained within the remit of the energy Council. In this case, the

73 See Annex II for the text of Rule 50 of the procedures of the European Parliament on associated committees.

132 CPI into EU renewable energy policy involvement of internal pro-climate stakeholders is closer to low according to Table 3.1.74

As an aggregate, however, the involvement of pro-climate stakeholders in the policy process leading to the adoption of the RE Directive is medium to high. The low levels of involvement of the environment formation in the Council on the RE Directive lowers the medium-high and high levels of involvement in the Commission and the Parliament.

External pro-climate stakeholders As discussed above, in the lead up to the publication of the proposal for the RE Directive, the Commission had several open and public online consultations related to specific issues around RE policy. NGOs and members of the RE industry were involved and present in these public consultations (ibid.; interview 5). Procedures were in place to allow them to provide input in the early stages of the drafting of the proposal. In each of the public consultations mentioned above, the proportion of environmental NGOs was generally small, due to large numbers of responses from many interested stakeholders (including member states, citizens, other industries etc.).75

In general, pro-climate external stakeholders were most concerned with the biofuels criteria, and with ensuring interim mandatory targets for renewables. There was thus some pessimism that the Commission’s proposal did not elaborate on either of these aspects, and that these points were not strengthened in the final output (ENDS Europe, 2008c, 2008e). However, the procedures for consultation allowed space for external pro-climate stakeholders to express their opinions. NGOs felt they could have access to the Commission if they wished it. As one interviewee mentioned, she was never refused a meeting with a Commission official to discuss the RE proposals between 2006 and 2008 (interview 7). The fact that the RE proposal was part of the overall climate and energy package meant that the general climate issue was often discussed at stakeholder meetings, and not necessarily with a focus on RE solely. Other RE

74 The European Council also played a role in the negotiations on the 2009 RE Directive, but as a predominantly political institution with no official co-deciding responsibilities (see chapter two), I discuss the role of the European Council as part of the explanation of political commitment later.

75 For example, the consultation on the Commission’s energy green paper in advance of the second strategic energy review attracted 164 written comments, of which 22 were from (all types of) NGOs (European Commission, 2006f).

133 CPI into EU renewable energy policy industrial actors also agreed that, at this time, there were no barriers to access to the Commission. The Commission did not necessarily seek out input of the pro-climate stakeholders, but they were open to any requests for meetings (interviews 5 & 7).

Therefore, in line with Table 3.1, the involvement of external pro-climate stakeholders in the policy process with the Commission leading up to the publication of the RE proposal is medium-high. Procedures were in place so that environmental NGOs and the RE industry could provide their opinions to the policy process. There was little opposition (but some opposition from major GHG-emitting industries, reticent member states, for example) to pro-climate stakeholder arguments. In addition, access to members of the Commission on this file was considered relatively easy.

In the Parliament, most interviewees agree that access for external stakeholders was easy (interviews 5, 6, 7, 8, 25 & 26), although this ease of access was (and is, in general) often linked to personal relationships with MEPs and/or their assistants. In the case of the negotiations around the RE Directive, with Green MEP Claude Turmes acting as rapporteur on this dossier, the Parliament was particularly open to input from pro-climate external stakeholders (Boasson & Wettestad, 2013; interview 7). The rapporteur also created a group of stakeholders (including members of the EREC organisation) to facilitate the drafting of the committee’s report, with a special focus on the issue of renewables trading (Boasson & Wettestad, 2013, p. 91). The Parliament was therefore an open institution for external pro-climate stakeholders, with the rapporteur (as a green MEP) being particularly open to opinions of environmental NGOs and the RE industry. Although the negotiations in the policy process (especially in the later stages) involved much informal negotiating behind closed doors, the contacts between external pro-climate stakeholders and Parliamentary actors were nonetheless considerable (interview 7). In sum, and according to the indicators described in Table 3.1, the level of involvement of external pro-climate stakeholders in the Parliament is high, with easy access to the policy process in the Parliament.

In the Council, it is generally noted that access for external pro-climate stakeholders is difficult (Hauser, 2011; interviews 5, 6, 7, 8 & 11). In the case of the RE Directive, however, there were certain member states that were more interested in strong and

134 CPI into EU renewable energy policy ambitious policy than others. Environmental NGOs described that they made efforts to target these member states, even in the years before the proposal was published (interview 7). With the Council, some of the conventional power industries had powerful lobbies at national level to support a system of tradable certificates that was seen as disadvantageous for RE development by environmental NGOs and the RE industry (Boasson & Wettestad, 2013; interview 7). Pro-climate stakeholders in this case had some success in targeting specific member states that shared their opinions (ibid.). Nevertheless, except on the national level and by targeting specific national governments, there was little to no access to the Council for external pro-climate stakeholders. Additionally, with the proposal being agreed in first reading, the process was rapid and much of the negotiating took place informally and behind closed doors. There was no room for external pro-climate stakeholders to take part. The level of involvement of external pro-climate stakeholders in the policy process in the Council is thus low (see Table 3.1).76

Taken together, the medium-high levels of involvement of external pro-climate stakeholders in the Commission and high levels in the Parliament and the low levels of involvement in the Council point overall to medium-high levels of involvement in the policy process overall. 77

Recognition of functional interrelations In the case of the policy process leading to the 2009 RE Directive, there are signs that the functional interrelations between climate policy objectives and RE policy were recognised by policymakers. However, the need for highly ambitious RE policy to achieve long-term climate policy objectives was less a part of the discussions.

Climate policy objectives and RE policy objectives to increase the share of RE interrelate harmoniously, as increasing the share of (most sources of) RE in energy consumption directly results in reductions of GHG emissions from energy, which helps achieve climate policy objectives. Since the agreement in March 2007 in the

76 The European Council is discussed below as part of understanding the levels of CPI in the 2009 RE Directive.

77 Generally, external pro-climate stakeholders were relatively positive about the final Directive. The European Wind Energy Association, EWEA, described the Directive as confirming “Europe as the leader of the energy revolution the world needs”; EREC described it as the “most important piece of legislation on renewable energy in the world”; Greenpeace described it as the “dawn for a clean energy future that will benefit the climate and the economy” (ENDS Europe, 2008e). Environmental NGOs were still critical of the lack of progress on sustainability criteria for the use of biofuels, however, with Greenpeace, the European Environmental Bureau (EEB) and Wetlands International pessimistic about the implications of the promotion of biofuels (ibid.).

135 CPI into EU renewable energy policy

European Council to move to a 20 per cent target for the share of RE in the EU by 2020, RE policy development was specifically linked to meeting climate policy objectives. In 2007, the European Council stated that it: “reaffirms the Community’s long-term commitment to the EU-wide development of renewable energies beyond 2010” and that it endorses “a binding target of a 20% share of renewable energies in overall EU energy consumption by 2020” (European Council, 2007, p. 21). In 2008, the European Council linked this target very clearly to its ambitions for providing leadership in the international climate negotiations, when it stated: “The EU is committed to maintaining international leadership on climate change and energy… By delivering on all the targets set by the spring 2007 European Council, the EU will make a major contribution to this objective” (European Council, 2008a, p. 11). While the European Council demonstrates high recognition of the interrelations between RE and climate policy (thus outlining the political lines for the various Council formations), at least in the demands for binding policy measures in RE to combat climate change (see Table 3.1), we see that throughout the policy process it is the Council that is the most reticent actor to strengthen policy measures further (see section 4.3.2 above).

In its proposal, the Commission also clearly outlined its recognition of the functional interrelations between climate policy objectives and RE policy objectives. The RE Directive was proposed by the Commission as part of the climate and energy package that aimed to achieve the so-called 20-20 targets by 2020. The opening sentence of the Commission’s proposal demonstrates this recognition: “The Community has long recognised the need to further promote renewable energy given that its exploitation contributes to climate change mitigation through the reduction of greenhouse gas emissions…” (European Commission, 2008d, p. 2). The Commission, in its proposals, stuck to the Council’s agreed 20 per cent target for RE share in 2020, although recognising the long-term requirements for RE on the road to decarbonisation ought perhaps to have increased the level of ambition. The Commission shows high recognition of the functional interrelations between RE policy and climate policy objectives, with RE policy proposals aiming to achieve (short- to medium-term) climate policy objectives (see Table 3.1).

The Parliament also clearly recognised the functional interrelations between achieving the objectives of climate policy and the objectives of RE policy. Of the three

136 CPI into EU renewable energy policy institutions, it was the most ambitious, in that it called for more ambition for 2020 (a target of 25 per cent RE share by 2020; European Parliament, 2007). The Parliament pushed for measures within the RE Directive to ensure the 20 per cent target would be achieved. Although, in the end, it compromised on a number of issues with the Council to ensure agreement was reached, the rapporteur linked the RE Directive throughout the negotiations to the credibility of the EU’s leadership on climate change internationally (ENDS Europe, 2008e). The Parliament demonstrates high recognition of the functional interrelations between RE policy and climate policy objectives throughout the policy process leading to the adoption of the 2009 RE Directive.

In line with Table 3.1, the recognition of the functional interrelations between RE policy and climate policy objectives was high in that achieving climate policy objectives is one of the main stated goals of RE policy, and this remained a part of the policy discourse throughout the process. The recognition could not be considered very high, however, as policymakers did not seek to be as ambitious on RE policy as long- term climate policy objectives would suggest.

4.3.2.2 CPI in the policy output of the 2009 RE Directive Figure 4.2 outlines the BAU scenario from 2011 that shows the policy output of the 2009 RE Directive, and compares that to very high or complete levels of CPI on a trajectory from 2000 to 2050. According to this graph, very high levels of CPI in 2020 would see shares of between 37 and 45 per cent of RE in the EU. BAU scenarios from 2006 suggest that the share of RE in the EU was expected to reach about twelve per cent by 2020. Thus, the 2009 RE Directive, with its target to reach a 20 per cent RE share, closes the gap (25 to 33 percentage points) between BAU and very high levels of CPI for 2020 by eight percentage points. This results in levels of CPI in the policy output of 24 to 32 per cent. According to Table 3.2, this puts the level of CPI in the policy output of the RE Directive at low levels.

However, given that the RE Directive was negotiated in 2008, the actual shares of RE in 2005 were known and could form the starting point for a CPI trajectory to 2050 – meaning that the BAU scenario changed between the negotiations on the 2001 RES-E Directive and the 2009 RE Directive, due mainly to deficits in the implementation of previous RE policy. The share of RE in 2005 was close to nine per cent (EEA, 2008).

137 CPI into EU renewable energy policy

Starting from 2005 levels, very high levels of CPI in the policy output would reach a share of RE in the EU of between 33 and 40 per cent (see Figure 4.5) in 2020. The 2009 RE Directive can then be considered to close the gap between BAU (twelve per cent share in 2020) and very high levels of CPI in 2020 (33 to 40 per cent share), by between 29 and 38 per cent share. This remains at (the higher end of) low levels of CPI in the policy output (see Table 3.2).

Figure 4.5: CPI in the policy output of the 2009 RE Directive, compared to BAU scenarios and the very high CPI trajectory from 2005 levels.

Source: EEA, 2008; European Commission 2011g; own calculations. Finally, the policy output could be measured in a way that tempers the results in the early stages on the path towards decarbonisation. As discussed above on the policy output of the 2001 RES-E Directive, there are some valid arguments to be made about the challenges of developing the RE share in a linear trajectory that makes no distinction between early, costly development, and speedier development in later decades as costs for installing RE fall. Figure 4.4 above suggested a midway point for a very high CPI target of a RE share in 2010 of between 17 and 19 per cent, which corresponds with 2010 levels in the CPI trajectory in Figure 4.5. Measuring from 2005, this implies an increase in the share of RE to 2050 of between eight and ten percentage points every five years. This already seems to be a significant increase, considering the history in the EU of achieving an increase of about one percentage point every five years since 1990. Nevertheless, it may be realistic to assume that RE shares to 2020 will increase at a slower pace than by eight to ten percentage points

138 CPI into EU renewable energy policy every five years, due to relatively high up-front costs for expanding RE.78 An example of reduced early development could rather see between six and seven percentage point increases every five years to 2020 – as a direct follow-on from the midway point outlined in Figure 4.4 (with six to seven percentage point increases every five years from 2000 to 2010). Figure 4.6 shows what such a trajectory would look like to 2020 (and beyond to 2050, with steeper RE share development required between 2020 and 2050). Such a scenario suggests that between nine and twelve percentage point increases in the share of RE would be required every five years from 2020 to 2050.

Figure 4.6: CPI trajectory for the RE share to 2050, with lower expectations to 2020.

Source: EEA, 2008; European Commission, 2011g; own calculations. Following the sort of trajectory outlined in Figure 4.6 implies very high CPI levels of between 27 and 30 per cent RE share by 2020. The output of the RE Directive of a 20 per cent share thus closes the gap between BAU (twelve per cent RE share in 2020) and very high levels of CPI by eight percentage points, with a score of between 44 and 53 per cent. This corresponds to medium levels of CPI in the policy output to 2020 for the RE Directive, if we assume that steeper penetration of RE in the overall share of EU energy can be more easily achieved after 2020 with the costs of RE technology dropping. Such a trajectory does, however, leave a substantial part of the work for the later decades leading to 2050, while climate science argues in favour of policies that ensure that GHG emissions peak as soon as possible (IPCC, 2007).

78 While it does not feature strongly in the analysis here, the economic crisis from 2008 onwards has resulted in lower-than- expected investments in RE (European Commission, 2013d).

139 CPI into EU renewable energy policy

Overall, the level of CPI in the policy output of the 2009 RE Directive can be considered low-medium, but closer to low if it is assumed that linear or early action is required to achieve climate policy objectives.

4.4 Explaining CPI in EU RE policy Table 4.2, below, summarises the results of the analysis. The Table shows little evolution in the extent of CPI in the policy output of EU RE policy between 2000 and 2010, although the level of CPI in the policy process over the same time period increased from low-medium to medium-high. The nature of the EU’s RE policy also developed from indicative targets in 2001 to mandatory targets in 2009.

CPI in the 2001 RES-E CPI in the 2009 RE

Directive Directive CPI in policy output Low Low (to medium) Internal pro- Low- Medium- climate medium high stakeholders: CPI in External pro- the Medium- climate Low Low Medium policy high stakeholders: to to high process Recognition medium of functional Medium High interrelations: Table 4.2: Levels of CPI in the policy process and output of the 2001 RES-E and the 2009 RE Directives. In this section I apply the four explanatory variables described in chapter three to the story of CPI in the 2001 RES-S and 2009 RE Directives to try to understand why such levels of CPI were found in each case. The four explanatory variables include functional interrelations; political commitment; the institutional and policy context (each of which can explain the level of CPI in the policy output and policy process); and the process dimension (which can explain the level of CPI in the policy output only). The first three variables are discussed below with the view that their influence in the process could lead to influence in the output (as output follows from process). Thus I do not discuss separately the effect of each variable on the level of CPI in the policy process and the policy output, as these two are linked (Briassoulis, 2005a).

140 CPI into EU renewable energy policy

4.4.1 Functional interrelations First, I examine the nature of the functional interrelations between EU RE policy objectives and long-term climate policy objectives. Where functional interrelations exist, some demand for CPI may occur. In addition, functional interrelations may be more or less direct or indirect (or closer or farther apart); and more or less synergistic or conflictual. The more direct and synergistic the functional interrelations, the more likely that higher levels of CPI can be expected as policy connections are harmonious and obvious (see Table 3.4, which shows a matrix of the nature of functional interrelations and their potential effect on the level of CPI).

The functional interrelations between RE policy and climate policy are direct and synergistic. The objectives of the 2001 RES-E Directive are to increase the share of RE in the EU to twelve per cent by 2010, corresponding to a 22 per cent increase in RE in electricity. For the 2009 RE Directive, the objective is to increase the share of RE in final EU energy consumption to 20 per cent by 2020. These objectives directly affect the long-term climate policy objectives in a synergistic way. In other words, increasing the share of (most sources of) RE reduces the level of GHG emissions by displacing fossil fuels. Scenarios to achieve long-term climate policy objectives of decreasing GHG emissions in the EU by between 80 and 95 per cent point to the highly important role to be played by RE in the future (EREC, 2010; European Commission, 2011a; Heaps, et al., 2009; WWF, 2011). Although the long-term climate policy target was not described in the same terms at the time of the negotiation of the 2001 RES-E Directive, the level of ambition required was nonetheless known. Limiting global temperature increase to 2°C (IPCC, 1995; European Council, 1996) – commonly agreed, since the 1990s, as the long-term climate policy target – requires clearly ambitious policy measures to reduce the emissions of GHGs. Developing RE policy contributes to achieving these goals.

Thus, the nature of the functional interrelations in this case is most favourable to promoting higher levels of CPI in the policy process and output of both the 2001 RES-E and 2009 RE Directives (see Table 3.4). The nature of the functional interrelations between long-term climate policy and RE policy is not something that changes or evolves, but is simply a reality of RE and climate policies.

141 CPI into EU renewable energy policy

4.4.2 Political commitment When discussing the explanatory role of political commitment to levels of CPI, I assess, first, the political commitment of the EU and its institutions to combating climate change in an overarching sense. Secondly, I assess levels of political commitment to promoting CPI in other policies (in this case, in RE policy).

Early development of RE policy in the EU in the 1970s and 1980s was rather at member state level and rather in response to concerns of energy security (especially in the wake of the oil crises of the 1970s; see chapter two; Hildingsson, et al., 2010). Climate change was not an issue of political concern in these early days, but as the issue rose on the international political agenda, the EU began to pay more attention to the potential of RE policy as a response to the climate challenge (Wurzel & Connelly, 2011b). RE policy was linked very clearly to the international negotiations on climate change, and the 2001 RES-E Directive was proposed (in 2000) as a policy that would help the EU meet its agreed commitments under the Kyoto Protocol (European Commission, 2000e). This was a period in which the EU expressed its desire to provide international leadership on climate change. After a decade of few credible domestic policies to back up the EU’s claims for leadership, the RES-E Directive could have provided some credibility to these claims. However, with no binding targets agreed in the RES-E Directive, and few other major policy proposals at EU- level at this time, this was not a peak period for demonstrated political commitment in the EU to climate change (Oberthür & Roche Kelly, 2008; Wurzel & Connelly, 2011b). The early 2000s was a period of symbolic, rather than concrete or credible, EU political commitment to combating climate change generally (see chapter two). Table 3.5 describes indicators for measuring the level of political commitment over a threefold scale (low, medium, high). In this case, political commitment to climate change generally is rather medium, with expressed commitment to combating climate change, but few credible policy measures to follow up.

By the mid- to late-2000s, overall political commitment in the EU to combating climate change had evolved. The 2009 RE Directive was proposed and negotiated in 2008. This period is considered the peak of political commitment to climate change generally in the EU (Eckersley, 2012; interview 22; Oberthür & Roche Kelly, 2008; Wurzel & Connelly, 2011b). The European Council (the highest political level of the

142 CPI into EU renewable energy policy

EU) had come out in March 2007 in favour of a binding target to reduce GHG emissions by 20 per cent by 2020, compared to 1990 levels; and a binding target to increase the share of RE in the EU to 20 per cent by 2020 (European Council, 2007). This period marks a shift in EU commitment to leadership on climate change from symbolic leadership based on rhetorical statements, to credible leadership with policy measures to move forward on combating climate change in the EU (Wurzel & Connelly, 2011b). The EU was particularly anxious at this time to demonstrate its credible leadership in advance of the 2009 international negotiations at the fifteenth Conference of the Parties (COP) to the UNFCCC in Copenhagen, and pushed for the adoption of domestic policy measures in advance of this meeting. At this point in time, political commitment to combating climate change generally was at a high level in the EU (see Table 3.5), with statements expressing the political commitment followed up with concrete policy measures.

Secondly, I assess political commitment to CPI leading to the adoption of EU RE policy. With regard to the 2001 RES-E Directive, the Commission justified its proposal with reference to the Kyoto Protocol commitment, and clearly saw RES-E policy mostly as a climate policy measure. In the proposal, the Commission explained in its first paragraph that the RES-E Directive would be “an important part of measures aimed at meeting the obligation to reduce the emission of greenhouse gases accepted by the EU at Kyoto…” (European Commission, 2000e, p. 2). The EU (the EU15 at the time) had agreed to reduce its GHG emissions by 8 per cent over 2008- 2012 compared to 1990 levels under the Kyoto Protocol. This appears to have been the main motivation in the Commission’s proposal for the Directive, although other motivations were also mentioned (including improving the security of energy supply, and social and economic cohesion; European Commission, 2000e, p. 3). Additionally, the Commission highlighted that the RES-E Directive helps achieve EU objectives to integrate environmental protection requirements into other policies (Art. 6, TEC; see Annex II). Nevertheless, these motivations did not push the Commission to propose a strong, ambitious Directive with mandatory targets. In this case, the Commission overall demonstrates medium-high levels of political commitment to CPI in the process leading up to, and the promotion of the policy output of, the 2001 RES-E Directive (with climate objectives providing primary motivations for the policy

143 CPI into EU renewable energy policy proposal, but insufficient commitment to CPI to motivate a stringent proposal with binding targets).

The Parliament’s proposed amendments aimed to strengthen the Commission’s proposal, especially by making targets mandatory (European Parliament, 2000b). In the explanatory statement accompanying the Parliament’s first reading report, the importance of increasing the share of RE is highlighted: “renewable energies are an integral feature of an effective strategy to protect the climate; they help achieve the Kyoto objectives…” (European Parliament, 2000b, p. 37). Throughout the negotiations, the rapporteur on the dossier (MEP Mechtild Rothe) continued to push for mandatory targets (ENDS Europe, 2000c, 2001c). In the end, she could only compromise to reach agreement with the Council. In the explanatory statement accompanying the second reading, the rapporteur was critical of the Council for not following through on commitments made internationally: “at Kyoto, the European Union accepted its responsibilities and committed itself to meeting certain targets… If we actually wish to meet those targets, and we should assume that we do, why are we not committing ourselves to making the necessary measures binding?” (European Parliament, 2001a, p. 15). In this case, and in line with Table 3.5, the European Parliament demonstrates high levels of political commitment to CPI in the 2001 RES- E Directive. Although the Parliament compromised in the end with a more reticent Council, the aim of the Parliament was to strengthen the Commission’s proposal to achieve the Kyoto Protocol climate objectives.

In the Council, the level of political commitment to CPI in the 2001 RES-E Directive is different to both the Commission and the Parliament. In the lead up to the Commission’s proposal, the Council pushed the Commission to propose legislation on RE (ENDS Europe, 1999b), but during the negotiations on the proposal, several member states lowered the overall ambition. In its common position on the Commission’s proposal, the Council reinforced the idea of indicative non-binding targets (Council of the European Union, 2001). It rounded the overall target for electricity to 22 per cent RE share (when Parliament suggested this should be rather 23.5 per cent, see section 4.3.1 above), and it broadened the scope of the definition of renewable energy sources to include large-scale hydropower, and waste incineration (ibid.). While the Council kept references to the importance of the RES-E Directive as a measure to achieve climate policy objectives under the Kyoto Protocol in the

144 CPI into EU renewable energy policy recitals, the negotiations themselves and the final Directive all indicate symbolic political commitment to CPI in RES-E policy (i.e. statements of commitment without following through with stringent policy measures). In this case, the level of political commitment in the Council to CPI for both the policy process and the policy output of the 2001 RES-E Directive is rather medium, as outlined in Table 3.5 (with evidence of expressed commitment to CPI).

In the later RE policy developments, it is clear that the proposal for an RE Directive in 2008 was considered by EU policymakers as a direct response to commitments to combat climate change. It was drafted after the European Council endorsed the 20 per cent target for RE share by 2020 in March 2007 (European Commission, 2008d, p. 2; European Council, 2007). The Commission stated in the proposal that: “the challenges of climate change caused by anthropogenic emissions of greenhouse gases, mainly from use of fossil energy, need to be tackled effectively and urgently” (European Commission, 2008d, p. 2). The Commission included the RE Directive proposal as part of the climate and energy package (see above). Each of the proposals in this package aimed primarily to agree on policy measures to combat climate change. The RE Directive is an energy policy that aims to achieve climate policy objectives. In this case, the level of political commitment in the Commission to advancing CPI is high (see the threefold scale in Table 3.5).

The same level of political commitment to advancing CPI can be found in the Parliament. In terms of advancing CPI, the Parliament aimed to do so by calling for more ambitious policy on RE. As RE policy is synergistically related to climate policy, advancing ambitious policy objectives in RE policy advances climate policy. In the Parliament, calls for higher targets for 2020 were made in 2007 already. In its resolution on climate change of 14 February 2007, the Parliament called for the EU to agree on a 30 per cent GHG emissions reduction target for 2020 and a target of increasing the share of RE in the EU to 25 per cent by 2020 (European Parliament, 2007). Throughout the negotiations on the 2009 RE Directive, the Parliament also attempted to push the Council to agree on interim targets to achieve the 2020 target and clear sanctions in case these targets were not met. In the end, Parliament did not have its way on this particular aspect of the proposal, but it can still be said that Parliament demonstrated high levels of political commitment to supporting CPI in the 2009 RE Directive.

145 CPI into EU renewable energy policy

In the Council, as in the European Council in 2007, it seems that political commitment to combating climate change by advancing CPI in energy policy existed at this time. The Presidencies in 2008 (Slovenia and France) were both interested in pushing forward on all the policy proposals in the climate and energy package. France, in particular, expressly aimed to achieve agreement within the term of its Presidency in the second half of 2008 (Boasson & Wettestad, 2013). While member states were uninterested in Parliament’s proposal for binding interim targets and sanctions, they did agree with Parliament that sustainability criteria would be required for biofuels (ENDS Europe, 2008e). The fact that the RE Directive was negotiated alongside the other measures in the climate and energy package also led to some member states cooperating more on RE to receive more advantageous measures in the other proposals. Poland, for example, was a cooperative and progressive party in the negotiations on the RE Directive in order to gain special treatment for its coal plants under the ETS negotiations (Boasson & Wettestad, 2013). In this case, although there were attempts to water down significantly other measures in the climate and energy package, the main ambitious elements of the 2009 RE Directive remained in place from the proposal to the agreed text – member states did not negotiate on the 20 per cent target or on its binding nature. The RE Directive was considered throughout the process as an appropriate measure to combat climate change and demonstrate EU leadership on the issue. Additionally, the Council was pushed by the highest political level in the European Council in the negotiations, where final agreement on the RE Directive was announced. Thus, in line with Table 3.5, the political commitment of the Council to advancing CPI in the 2009 RE Directive is also rather high.

In summary, the level of political commitment to climate change generally in the EU evolved from a medium to a high level between 2001 and 2009. Political commitment to CPI measured medium-high in the policy process and output of the 2001 RES-E Directive, as an aggregate result of the medium-high political commitment to CPI demonstrated by the Commission; high in the Parliament; and, medium in the Council. Political commitment to CPI in the 2009 RE Directive amounts to high on aggregate, with high levels demonstrated by the Commission, Parliament and Council (and European Council) in the policy process and output at this time.

146 CPI into EU renewable energy policy

4.4.3 Institutional and policy context There are several internal and external institutional and policy contextual factors that may also provide some explanation for the levels of CPI found in the policy processes and outputs of the 2001 RES-E and 2009 RE Directives (see Table 4.2).

First, the institutional context at the time of negotiations may shed some light on why such levels of CPI were found. Both Directives were agreed under the co-decision procedure (today known as the ordinary legislative procedure). The legal basis was finally set at Article 175(1) (environment) of the TEC for the 2001 RES-E Directive, and Articles 175(1) and 95 (internal market) for the 2009 RE Directive (Council of the European Union, 2001; see Annex II). Under the co-decision procedure, negotiations took place between the Parliament and the Council, and a qualified majority voting (QMV) rule was applicable in the Council. With QMV, no single member state could veto agreement on the policy proposal. In the case of the 2001 RES-E Directive, individual member state concerns were worked out during negotiations and the voting rule in itself did not obviously push for policy change. For the 2009 RE Directive, the Council (pushed by the French Presidency especially) explicitly aimed for unanimous support of the agreement, although the voting rules allowed QMV (ENDS Europe, 2008k). The Council continued negotiations until all member states were satisfied with the result rather than having member states vote against a proposal. In these two cases, the QMV rule did not support action for more ambitious policymaking, as Council was rather unified in its unambitious approach to the RES-E proposal. The QMV rule exhibits neither a favourable nor unfavourable institutional context for advancing CPI for the 2001 RES-E Directive, but slightly more unfavourable for the 2009 RE Directive. As the QMV rule was not applied in the process leading to the 2009 RE Directive, the decision-making procedure may have been more in favour of reaching compromises and less in favour of policy change. Thus, this is more unfavourable to advancing CPI in RE policy in 2008.

The consultation procedures also evolved over time. For the 2001 RES-E proposal there were no standard Commission procedures for carrying out impact assessments and public consultations. This meant there were few official procedures for external stakeholders to get involved in the early stages of the policy process. These standards were adopted as part of the later better regulation push in 2002 (European

147 CPI into EU renewable energy policy

Commission, 2002d), and were applied in the 2009 RE Directive. Stakeholder involvement followed usual lobbying activities in both cases, but for the 2009 RE Directive there was the advantage of official consultations. This contextual factor is more unfavourable for advancing CPI at the time of the 2001 RES-E Directive.

In both the 2001 RES-E Directive and the 2009 RE Directive, there is evidence that policy developments continued along the same development path, and followed previous policy attempts rather than deviating radically from past policy paths. The 2001 RES-E proposal came in the wake of the Commission’s 1997 white paper that resulted in difficult discussions (Boasson & Wettestad, 2013; European Commission, 1997b). The Commission compromised on the issue of harmonisation of support schemes and proposed simply to monitor national support schemes instead (European Commission, 2000e). This demonstrates an interesting level of pragmatism in the Commission, which continued to propose measures despite historical weaknesses. Its earlier proposals in the 1990s, such as the ALTENER programme, were substantially weakened in the later policy negotiations. This past slow policy context could be seen as favourable for advancing policy further and advancing CPI, although this is softened with the difficult discussions in advance of the proposal. The expected failure that the EU would meet its 2010 targets for RES-E then helped push for more ambitious policy with mandatory targets for the 2009 RE Directive. The Commission’s 2007 renewable energy roadmap suggested a new policy framework including binding targets and using one policy instrument for overall EU energy consumption as a response to the disappointing implementation of the 2001 RES-E Directive (European Commission, 2007b, p. 4). This past policy failure was a motivation for proposing binding targets for RE shares to 2020. This contextual factor provides a more favourable context for advancing levels of CPI in RE policy. Thus, in both the RES-E and RE Directives, past policy failures or inadequacies provided a favourable context for advancing more ambitious policy and thus pushing for more CPI.79

79 This contextual factor also has an unfavourable aspect in that the failure of past policy also leads to constant “catch-up” governance, where future policy developments are trying to right the failings of policies that should have brought society to a certain point. Additionally, following on from past policy developments with the same framing allows little space for radical departures from previous policies (i.e. radical policy change is difficult).

148 CPI into EU renewable energy policy

In the case of the 2009 RE Directive, there was also the peculiarity that it was proposed as part of a package of policy measures. Member states especially, linked negotiations on the RE Directive with the other policy proposals in the early stages of the negotiations (see above the example of Poland). In the later stages of negotiations, this was less the case as most of the major issues on the RE Directive were resolved relatively early in the negotiations compared to the process in the other policies of the package (ENDS Europe, 2008f, 2008l). However, what is more particular about the RE Directive (and the entire climate and energy package) is the speed with which agreement was reached, and the fact that the dossier was concluded in the first reading. From the time of the proposal of the package (23 January 2008) to the agreement on the RE Directive (4 December 2008), just over ten months had elapsed. The speed of the negotiations, and the fact that these negotiations were proceeding alongside negotiations on the other elements of the package, meant that issues were resolved and compromises found quickly (interview 7). The speedy process is linked to the role of the international climate change negotiation agenda (see below). This contextual factor is more likely to be favourable to advancing CPI in RE policy, although it could be argued that the speed of negotiations meant that some issues were compromised upon too quickly (interview 6).

Finally, the role of the European Council and of the Presidency, as two overarching guiding actors in the EU is an additional institutional contextual factor. In the 2001 RES-E Directive, member states were generally reticent on pushing for stringent, binding targets, with the European Council and Presidency playing no special role to remedy this. Thus, for the 2001 RES-E Directive, this context plays no role in explaining the levels of CPI found. In the process leading to the 2009 RE Directive, however, the Presidency and the European Council both played strong roles in pushing for agreement on the 2009 RE Directive. The French Presidency, especially, aimed to reach agreement on the entire climate and energy package before the end of its mandate (ENDS Europe, 2008d). In the end, political agreement on the climate and energy package was reached in the European Council on 12 December 2008 before the Parliament voted on the package in its plenary (on 17 December 2008; ENDS Europe, 2008a, 2008k). As with the discussion on the package and speed of the agreement of the climate and energy package, the role of the European Council and of the Parliament is more likely to be favourable to advancing CPI, but this is difficult to

149 CPI into EU renewable energy policy establish. It could be favourable in terms of pushing policy developments forward, but unfavourable in terms of reaching too much compromise for the sake of a speedy agreement (interview 6).

Second, there are several external policy context factors that may also shed light on the levels of CPI found in the 2001 RES-E and 2009 RE Directives. The international climate negotiation schedule is important in both the 2001 RES-E and 2009 RE Directives. In 2000, when the RES-E proposal was published, the EU had made commitments to reduce its GHG emissions by eight per cent on average between 2008 and 2012 compared to 1990, under the Kyoto Protocol. The RES-E Directive was intended as an instrument to meet these objectives, but also to play the role of showing the international community the EU’s leadership on climate change. The withdrawal of the US from the Kyoto Protocol in April 2001 occurred in advance of the second reading on the RES-E Directive proposal. The EU decided to pursue its ratification of the Kyoto Protocol, even without US involvement (Oberthür & Roche Kelly, 2008). These ambitions (though symbolic) for leadership in the international negotiations at this time are generally favourable to advancing CPI in domestic policy measures in the EU. The climate and energy package, including the 2009 RE Directive, was developed in response to climate change, but also to provide credible policy measures to back up the leadership claims of the EU in the international climate negotiations. By agreeing on binding commitments for itself to increase the share of RE and reduce its GHG emissions, the EU hoped to provide leadership by example in advance of the 2009 COP in Copenhagen. This COP was supposed to result in an agreement on the international legal framework for combating climate change after the time frame of the Kyoto Protocol (i.e. after 2012). This international negotiation schedule is likely to have sped up the negotiation process at the EU-level, but it does not seem to have pushed for even more ambitious policy (Boasson & Wettestad, 2013; interview 6). At this time, the international negotiating context can be both favourable and unfavourable to advancing CPI – favourable in terms of pushing agreement forward, but unfavourable in terms of giving into compromises in the name of a quick agreement.

For the 2009 RE Directive, the external policy context included the global financial and economic crises that hit in the autumn of 2008. Although the negotiations on the RE Directive were completed not long after that, and although several of the most

150 CPI into EU renewable energy policy pressing issues may have been agreed upon already, the level of ambition and the strength of the Directive may have suffered in the late stages of negotiations as a result of the crises. The review clause inserted at the end of the negotiations on the RE Directive is seen as one element that may be directly linked to the crises (ENDS Europe, 2008c). This factor was rather more unfavourable to advancing the levels of CPI in the RE Directive.

In sum, the institutional and policy context did not seem to push for more ambition on the 2001 RES-E Directive, but the context was nevertheless slightly more favourable to further CPI. In the 2009 RE Directive, the context was rather neutral. Neither favourable nor unfavourable contextual factors hold a majority in this case for explaining the levels of CPI found in the 2009 RE Directive. For both the 2001 and 2009 Directives, there seems to be no single internal or external institutional or policy context that can sufficiently explain the levels of CPI in both the process and the output of the two Directives examined here. Thus, there is little evidence to show that the context was a driving factor for CPI in the policy output and process, but these can be considered background factors that may provide added explanatory value to some of the other variables.

4.4.4 Process dimension In policymaking, the policy output is preceded by a policymaking process that involves negotiations and discussions on a particular issue to reach agreement on a policy measure. Thus, for an analysis of CPI in a particular policy, it is clear that there are links between the level of CPI in the policy process and the output – the policy output follows on from the policy process. As an explanatory factor, the process dimension can aid in explaining levels of CPI in the policy output only.

In the policy process leading to the 2001 RES-E Directive, the level of CPI was found to be low-medium (but closer to low; see Table 4.2). This level is an aggregate of the low-medium involvement of internal pro-climate stakeholders; low involvement of external pro-climate stakeholders; and low-medium recognition of the functional interrelations between RE policy objectives and long-term climate policy objectives. CPI in the policy output was also found to be low-medium. In the policy process leading to the agreement on the 2009 RE Directive, the level of CPI was found to be medium-high (see Table 4.2). This is an aggregate of the medium-high level of

151 CPI into EU renewable energy policy involvement of internal pro-climate stakeholders in the EU institutions in this policy process; the medium-high involvement of external pro-climate stakeholders, including environmental NGOs and RE industry representatives; and, the high recognition of the functional interrelations between RE policy objectives and climate policy objectives in the policy process. CPI in the policy output of the 2009 RE Directive was low (-medium).

As discussed also in chapter three, some sort of correlation between the level of CPI in the policy process and the policy output can be expected in most cases. It is unlikely (but not impossible) that CPI in the policy output would be much greater than the CPI found in the policy process. Or conversely, it could be argued that the higher the levels of CPI in the policy process, the more chances that higher levels of CPI could also be found in the policy output (Briassoulis, 2005). In the case of the 2001 RES-E Directive, the levels of CPI in both the policy process and output are similar, indicating a correlation that the output did indeed follow on from the process. In the 2009 RE Directive, there is a gap between the medium-high level of CPI in the policy process and the low (-medium) level of CPI in the policy output, which points to an idea that the process was insufficient to advance higher levels of CPI in the policy output. However, to understand the influence of the process dimension, this variable ought to be broken down into its various components.

For the 2001 RES-E Directive, the low levels of involvement of external pro-climate stakeholders stand out. There was little involvement of external pro-climate stakeholders beyond informal lobbying activities. The Commission had no open public consultation procedure at this stage. Many of the later negotiations on the policy proposal took place in informal negotiations behind closed doors. The recognition of the functional interrelations in this case, also, focused predominantly on the short-term climate policy objectives of achieving the EU’s commitments under the Kyoto Protocol. Perhaps the low involvement of external pro-climate stakeholders reinforced the lack of recognition of long-term climate objectives. Poor access and no procedures for the involvement of external pro-climate stakeholders meant they could not further highlight the functional interrelations between ambitious RE policy and the long-term climate policy objectives.

152 CPI into EU renewable energy policy

In the case of the 2009 RE Directive, breaking down the level of CPI in the policy process shows that for the Council, the interaction of internal and external pro-climate stakeholders was low. In fact, the Council may have been the main reticent actor in the policy process, but the low result of its openness to pro-climate stakeholders was bolstered in the measurement of CPI in the policy process by the involvement of these actors in the Commission and Parliament. It may be that the Council played a key role in weakening the level of CPI in the final policy output below a level that could have been expected based on the level of CPI in the policy process overall. Thus, in this case, breaking down the process dimension variable emphasises the role of the Council for understanding the low (-medium) levels of CPI found in the 2009 RE Directive.

4.4.5 Overview of explanatory framework Table 4.3 provides an overview of the role of each of the explanatory variables in understanding the levels of CPI in the policy output and process of both the 2001 RES-E and the 2009 RE Directives. CPI in the policy output of the 2001 RES-E Directive was found to be low, while CPI in the policy process was low-medium (see Table 4.2). The level of CPI in the policy output did not improve much to 2009 – from low to low (to medium), even with the target moving from non-binding to binding – but the level of CPI in the policy process in the 2009 RE Directive advanced over time to medium-high. Which explanatory variables are most appropriate for explaining these variances in the level of CPI in the policy process over time, but the similar results for the level of CPI in the policy output?

As shown in Table 4.3, the nature of the functional interrelations between RE policy and long-term climate policy objectives is overarching both time phases. The functional interrelations are direct (i.e., there are clear, obvious and minimal steps between the interrelations of RE policy objectives and climate policy objectives) and synergistic (the objective of increasing the share of (most sources of) RE interrelates harmoniously with the objective of climate policy to reduce GHG emissions). Political commitment to both climate change and advancing CPI in RE policy increased between 2001 and 2009. The institutional and policy context played varying roles in 2001 and 2009, with the effect being rather neutral in 2009, and slightly more favourable to CPI in 2001. The level of CPI in the process dimension was in line with

153 CPI into EU renewable energy policy findings for the level of CPI in the policy output in 2001, but there was a gap between the level of CPI in the policy process in 2009 and the level found in the policy output. In 2001 there was little involvement of pro-climate stakeholders in the policy process, and no guaranteed access (through procedures) for them in the co-deciding institutions. In 2009, procedures were in place guaranteeing access for external stakeholders in most of the institutions, but again external stakeholders complained about the Council as an opaque institution that is difficult to influence. Combined with overarching political commitment to climate change in this time phase, there was ease for internal pro-climate stakeholders to include their message in the policy process. In sum, it seems that, in both time phases, there were variables that were in favour of advancing levels of CPI and variables that were playing against the advancement of CPI. The greatest differences between the two time phases, however, lie in the levels of political commitment and in the levels of CPI in the policy process.

Explanatory variables 2001 RES-E Directive 2009 RE Directive 1. Functional interrelations Direct & synergistic Direct & synergistic (process and output) 2. Political commitment (process and output) - To climate change Medium High

- To CPI Medium-high High

3. Institutional and policy Slightly more context Neutral favourable (process and output) Low guaranteed involvement of pro- 4. Process dimension Crucial role of the climate stakeholders Council (output only) across co-deciding institutions Table 4.3: Summary of explanatory variables for CPI in the 2001 RES-E Directive and the 2009 RE Directive.

From the above, therefore, it can be said that the two crucial explanatory variables in the case of RE policy include the increasing levels of political commitment over time and the improvements in the involvement of pro-climate stakeholders in the policy process. While political commitment in 2001 was medium and medium-high, the level of CPI in the policy process remained at a low-medium level. In 2009, political

154 CPI into EU renewable energy policy commitment was high for both examined elements, and it is plausible that this commitment pushed the level of CPI in the policy process to the medium-high levels found. This is particularly evidenced in the fact that there was high recognition of the functional interrelations in the 2009 policy process, and medium-high involvement of pro-climate stakeholders. In this case, the recognition of these interrelations and the high levels of political commitment may have reinforced each other to result in medium-high levels of CPI in the policy process in 2009.

These two variables (political commitment and CPI in the policy process) are, although crucial, nonetheless insufficient for a more detailed understanding of the limited improvement in the level of CPI in the policy output over time. Further nuance can be found in a closer examination of the variables and their interrelations. The nature of the functional interrelations allowed for a clear linking of RE policy objectives and climate policy objectives, and ensured the recognition and articulation of these interrelations in the policy process. Breaking down the level of CPI in the policy process reveals the crucial role played by the Council in both time phases. In the 2001 RES-E Directive, the Council was reticent to agree on stringent policy although the Parliament attempted to push for this. The same can be said of the Council in the 2009 RE Directive, although the surrounding contextual factors at this time meant the Council stuck to the proposed binding target for 2020 (for example, the facts that the European Council agreed to the 20 per cent target in 2007, and that the EU, generally, aimed to demonstrate leadership on climate change internationally). The institutional and policy contextual factors also help reveal more subtle explanations, without seeming to play a crucial explanatory role. Moving from an indicative to a binding target over time is one of the main advances in EU RE policy between 2000 and 2010, yet this advance is tempered (from a CPI perspective) by the contextual factor of failing to achieve earlier targets and policy commitments. Such a situation means that later policy was being developed against a lower-than-anticipated status quo, leading to a situation of catch-up governance – while the later policy developments can be more ambitious, the failure of previous policies means that much of the ambition remains “too little, too late”. That later policy development took place in the context of catching-up on previous policy failures also highlights the lack of the long-term perspective in the policy discussions. The 2009 RE Directive policy output was both binding and more ambitious than previous policy, but the failure of

155 CPI into EU renewable energy policy the previous policy meant that even this output is insufficient from a CPI perspective. In this case, it seems the long-term perspective of achieving climate policy objectives to 2050, which should have suggested higher ambitions of RE policy in both time phases, did not feature strongly in the policy process.

In summary, in the case of RE policy, the two most important explanatory variables appear to be the levels of political commitment to both climate change and CPI combined with the level of CPI in the policy process (the process dimension). The nature of the functional interrelations and the surrounding institutional and policy contexts displayed a background context that was either neutral or favourable to the advancement of CPI. Political commitment and the process dimension then explained the levels of CPI in the policy process and output of the RE policy over time against this background.

4.5 Conclusions In this chapter, I examined the level of CPI in the policy output and process of the 2001 RES-E and 2009 RE Directives. The 2001 RES-E Directive aimed to increase the share of RE in EU energy consumption to twelve per cent by 2010 (corresponding to an increase in the share of RE in electricity generation to about 22 per cent by 2010). The targets agreed under the 2001 RES-E Directive were indicative targets, and ex ante reports indicate that they were not met (EEA, 2008; European Commission, 2006d). The 2001 Directive was proposed with the main motivation of achieving part of the EU’s commitment to reduce its GHG emissions under the Kyoto Protocol (although the objective to improve energy security was also promoted). The 2009 RE Directive was proposed in 2008 as part of the climate and energy package. It provided the legislative framework for the EU to meet the target of increasing the share of RE in EU energy consumption to 20 per cent by 2020, as endorsed by the European Council in March 2007 (European Council, 2007). Unlike the targets agreed under the 2001 RES-E Directive, the 2009 Directive imposed binding targets on member states to achieve the overall 20 per cent target.

Both the 2001 and 2009 Directives were proposed with part of their aims being the achievement of climate policy objectives. In the case of the 2001 RES-E Directive, this was framed in terms of contributing to achieve the EU’s emission reduction commitments under the Kyoto Protocol, to reduce GHG emissions by eight per cent

156 CPI into EU renewable energy policy between 2008 and 2012 compared to 1990 levels. This can be considered a short-term climate policy objective. The longer-term climate policy objectives (to meet the 2050 goals or to limit global temperature increase to 2°C) did not seem to play a major role in policy discussions leading to the adoption of the RES-E Directive. For the 2009 RE Directive, the EU aimed to demonstrate through unilateral action that it was a credible leader on climate change. It agreed the 2009 RE Directive in the first reading, motivated to influence international climate negotiations in 2009. The EU RE policy therefore has been closely linked to policy objectives on climate change between 2000 and 2010. At first glance, then, RE policy can be considered a case of well- integrated climate and energy policy.

As the analysis in this chapter revealed, however, the levels of CPI in EU RE policy are far from ideal in the policy output, when measured against benchmarks that aim to achieve the goals of long-term climate policy (namely, reducing GHG emissions in the EU by between 80 and 95 per cent by 2050). In both the 2001 and 2009 Directives, CPI in the policy output was, at best, low (-medium). There was limited improvement over the course of the decade, and the poor implementation of the 2001 RES-E Directive meant later policy developments were catching-up on potential improvements that did not materialise. The level of CPI in the policy process was low-medium in the 2001 Directive, but this level had increased to medium-high during the negotiations on the 2009 Directive, with more guarantees through procedures for the involvement of pro-climate stakeholders in the process.

Attempting to explain these levels of CPI, I assessed the role played by the four explanatory variables outlined in chapter three, namely: the nature of the functional interrelations between RE policy and climate policy; political commitment to combating climate change and to advancing CPI; the institutional and policy context, each of which explain CPI in both the policy process and policy output; and, the process dimension that explains the level of CPI in the policy output only. The major differences between the two phases of policy development could be found in the levels of political commitment and in the process dimension. Interestingly, in the case of the 2009 Directive, the higher levels of CPI in the policy process did not seem to push higher levels of CPI in the policy output. The levels of political commitment in the 2009 Directive may have influenced the development in the level of CPI in the process, but further background contexts can be found when the process dimension

157 CPI into EU renewable energy policy variable is broken down. In so doing, the crucial role of the Council is revealed as a more reticent actor in the co-deciding process, with a lack of procedural guarantees for allowing access for external pro-climate stakeholders.

These findings back up some of the literature on EPI that emphasises the importance of political commitment for the advancement of EPI or CPI (Jordan & Lenschow, 2010), but it seems clear from this case study that this is an insufficient factor for advancing EPI/CPI. Two interesting background contexts help provide nuance to the explanation – first, the earlier policy failures (likely linked to the indicative nature of the targets) that resulted in “too little, too late” in later policy developments, which meant that the recognition of the long-term climate policy objectives was insufficient to push for even more ambitious policy. Second, the process dimension also revealed the particular role played by the Council in both time phases for lowering the ambition or not providing space for pro-climate stakeholders.

EU RE policy therefore can be said to demonstrate insufficient levels of CPI in both the policy process and output between 2000 and 2010. The explanatory variables applied here highlighted the background contexts that may have resulted in this insufficiency, with two explanatory variables highlighted as having evolved over time and influenced the limited evolution in the levels of CPI between the 2001 RES-E Directive and the 2009 RE Directive. This same framework is applied, in the next chapter, to the EU’s energy performance of buildings policy.

158

CPI into EU energy performance of buildings policy

In this chapter, I discuss the second case study – EU policy on the energy performance of buildings. While chapter four on the EU’s renewable energy policy dealt with the domestic supply side of energy, I turn attention here to policies to manage and reduce energy consumption. With buildings representing the largest consumer of energy in the EU (followed by transport and industry; European Commission, 2013b, p. 6), any policy to reduce energy consumption must deal with the buildings sector.

Improving energy efficiency can be considered a win-win solution for a wide range of issues, including the goals of EU energy policy more broadly: security of energy supply, competitiveness and sustainable use of energy. Defined as “output divided by total (fuel) input”, or as actions that are designed to “save fuels”,80 energy efficiency involves the reduction in the consumption of energy. This implies reducing energy consumption of products, buildings, transport, power generation and heating. While several actions can be taken at the demand side (e.g. consumers choose the most efficient products and appliances, reduce their home heating temperature and electricity consumption), other actions are stimulated at the supply side (e.g. efficiency measures in power generation).

This chapter focuses on analysing the level of CPI in the policy processes and outputs of legislative measures to improve the energy performance of buildings. As a major consumer of energy (with buildings accounting for more than 40 per cent of final energy consumption in the EU; Andaloro, Salomone, Ioppolo, & Andaloro, 2010; BPIE, 2011; European Commission, 1984, 1998, 2013c), targeting their energy performance holds great potential for reducing energy demand and thus reducing demand for fossil fuels. I open the chapter with a discussion on EU energy efficiency policy over time. In the second section, I outline the benchmark for very high or complete levels of CPI that could be expected in energy performance of buildings

80 See the glossary available on eea.europa.eu, accessed: 23 September 2012.

159 CPI into EU energy performance of buildings policy policy. Third, I turn to EU policy developments in this sector between 2000 and 2010. I analyse the level of CPI in the policy output and process of the 2002 energy performance of buildings Directive (EPBD, Directive 2002/91/EC) and its 2010 recast (Directive 2010/31/EU). In section four, I explain the levels of CPI found with reference to the explanatory framework outlined and discussed in chapter three, before concluding the chapter.

5.1 Energy efficiency in the EU Energy efficiency is linked to energy consumption – the greater the improvements in energy efficiency, the less energy is consumed. The aim in promoting energy efficiency policy measures is precisely to reduce energy consumption, thus reducing dependence on imports of energy, reducing GHG emissions by generating less energy from fossil fuel sources, and reducing energy bills. Energy efficiency is often measured in terms of the “energy intensity” of an economy. The less energy intensive an economy (i.e. the less energy consumed for a fixed cost), the more efficient the economy is. As Figure 5.1 shows, the energy efficiency of the EU27 has improved between 2001 and 2010 (or, in other words, the energy intensity of the EU27 has declined between 2001 and 2010).

Figure 5.1: Development of energy intensity of the economy of the EU27 between 2001 and 2010, measured in kilograms of oil equivalent per EUR 1000, with (dotted) trend line.

Source: Eurostat database, epp.eurostat.ec.europa.eu.

160 CPI into EU energy performance of buildings policy

Levels of energy consumption, however, have not made the same improvements. Figure 5.2 shows that the final energy consumption in the EU27 has fluctuated between 111 and 119 Mega tonnes of oil equivalent (Mtoe) between 2000 and 2010. The biggest drop in energy consumption in the EU in that decade occurred between 2008 and 2009, with the onset of the economic crisis, but energy consumption increased again between 2009 and 2010. It is clear, therefore, that although energy efficiency has improved overall in the EU, these improvements in efficiency have been compensated by increases in the energy consumption generally – leading to a balanced picture of rather steady (or even slightly rising) levels of energy consumption.

Figure 5.2: Final energy consumption in the EU27 from 2000 to 2010, measured in Mega tonnes of oil equivalent (Mtoe), with (dotted) trend line.

Source: Eurostat database, epp.eurostat.ec.europa.eu.

These figures demonstrate one of the major difficulties with energy efficiency measures, known as the rebound effect. As defined by the European Environment Agency (EEA), the rebound effect “refers to behavioural changes or other systemic responses that can partly or fully offset the beneficial environmental effects of … new technologies” (EEA, 2012, p. 22). An example of direct rebound effect is that as energy efficiency measures lead to reduction in energy bills, consumers use the saved money to buy even more or bigger energy-consuming appliances. The rebound effect may have economy-wide equilibrium effects: as efficiency improves, consumption increases, thus resulting in no net benefits for the economy. According to Figures 5.1

161 CPI into EU energy performance of buildings policy and 5.2 this seems to be the case in the EU. However, other factors also play a role, such as increases in income and increases in population. For policymakers, promoting behavioural change and taking account of this “rebound effect” represent challenges when designing measures to improve energy efficiency.

The EU has put in place an array of policy measures aimed at improving energy efficiency. Table 5.1 lists some of the main legislative measures adopted, dating back to the late 1980s. As far back as 16 September 1986, the Council set the objective of improving energy efficiency by 20 per cent by 1995 compared to 1985 levels (Council of the European Union, 1986). At this time, it was clear that environmental or climate objectives did not play any role in motivating energy efficiency policy. The Council’s 1986 resolution listed several reasons for the promotion of energy efficiency measures, and highlighted energy security (especially in the context of rising EU dependence on oil imports) as a main motivation for improving energy efficiency. Improving the environment and combating climate change are not mentioned in this early resolution.

The Council’s position in support of energy efficiency for energy security reasons is a direct consequence and follow-on from the oil crises of the 1970s. Energy efficiency became an issue of high importance for the EU at this time. Improving energy efficiency and thus reducing demand for energy was seen as an effective response to concerns about growing dependence on foreign energy imports (especially oil), and thus an effective policy measure for promoting energy security in the EU (Boasson & Wettestad, 2013; European Commission, 1979a, 1984). Several member states responded with policy measures aimed at improving energy efficiency in their building sectors. National level energy performance standards for new buildings were put in place in the Netherlands, Denmark, Germany and France (European Commission, 1979b, p. 2). The Commission’s report on the EU’s early programme for energy saving in the 1970s already highlighted the future strategy for improving energy efficiency that later policy would follow, stating that “future savings will increasingly require investment in new equipment or buildings, or retrofitting the old, and more energy conscious behaviour from both investors and consumers” (ibid., p. 1).

162 CPI into EU energy performance of buildings policy

The Commission continued to suggest energy efficiency measures in the 1980s. It followed on from the focus on buildings in its 1979 report with a proposal entitled “towards a European policy for the rational use of energy in the building sector” in 1984 (European Commission, 1984). This document highlighted the great share in the energy consumption of buildings, stating “the domestic and tertiary sector accounts for 38% of Community’s energy consumption”, as well as a very significant saving potential of approximately 50 per cent of the total saving potential of all sectors (p. 1). The strategy for improving the energy performance of buildings was laid out in this document to include: the promotion of thermal auditing of buildings; technical improvements and regulations; use of funds for supporting efficiency measures; consumer/energy-user information. Much of the future regulation on energy efficiency, whether for buildings or appliances, incorporated these same themes. It was not until the late 1980s, however, that concrete policy measures at the EU-level were introduced (and not only discussions of targets for member states to meet with their own national measures). The 1989 construction products Directive was a first move at the EU level into building sector legislation (see Table 5.1).

As the energy efficiency issue moved down the political agenda of member states, with the memory of the 1970s oil crises fading, the Commission continued to propose (indicative) targets and policies in energy efficiency to enhance energy security. The Commission often highlighted the lack of progress in the EU on energy efficiency. In the proposal for the energy labelling Directive (European Commission, 1991), the Commission highlighted that the objective to improve energy efficiency by 20 per cent by 1995 compared to 1985 “was unlikely to be attained unless stronger policy measures were taken” (European Commission, 1991, p. 3). The 1992 energy labelling Directive would be the first major piece of legislation in the EU aimed to encourage energy efficiency through targeting energy conservation at the demand side (the consumer). Since then, several other pieces of legislation have been developed and redeveloped at EU level to improve energy efficiency (see Table 5.1).

Much of the EU’s energy efficiency policy in the 1990s was considerably watered- down from the Commission’s original proposal by member states during the ensuing policy negotiations.81 Member states were unconvinced about the importance of

81 For example, the SAVE Directive was substantially weakened in the policy process, with its proposed budget being slashed (Oberthür and Roche Kelly, 2008).

163 CPI into EU energy performance of buildings policy energy efficiency policy measures, and discussions around subsidiarity surfaced repeatedly when it came to advancing EU-level policy on efficiency. Later policy failures were attributed to inadequate implementation by member states (Boasson & Wettestad, 2013; Henningsen, 2011).

Year Legislation Code 1989 Construction products Directive 89/106/EEC 1992 Energy labelling Directive 92/75/EEC 1992 Efficiency for hot-water boilers Directive 92/42/EEC 1993 SAVE Directive 93/76/EEC 2002 Energy performance of buildings Directive 2002/91/EC 2004 Promotion of cogeneration Directive 2004/8/EC 2005 Ecodesign Directive 2005/32/EEC 2006 Energy services Directive 2006/32/EC 2009 Ecodesign Directive – recast 2009/125/EC 2010 Energy labelling Directive – recast 2010/30/EU 2010 Energy performance of buildings Directive – recast 2010/31/EU 2012 Energy efficiency Directive 2012/27/EU Table 5.1: Main pieces of EU legislation promoting energy efficiency.

The Commission began linking energy efficiency policy measures to the need to combat climate change in the 1990s. Boasson and Wettestad (2013) describe the international climate negotiations as representing an opportunity for the Commission to push again for ambitious energy efficiency measures internally in the EU (p. 152). As such, climate change became an added motivation for pursuing energy efficiency policy alongside energy security motivations. In the 1990s, measures to improve energy efficiency were touted by the Commission as necessary for the EU to achieve its target to stabilise its GHG emissions at 1990 levels (see, e.g., European Commission, 1992a, 1998). Since then, energy efficiency policy measures have come to be considered as climate policies as well as energy policies, and, on the surface, or at first glance at least, could be considered as excellent examples of climate policy integration into energy policy. It is certainly difficult to conceive of efficiency measures that could be bad for combating climate change. With the energy sector responsible for 80 per cent of GHG emissions in the EU (CANEurope, 2011, p. 14; EEA, 2010, p. 31; European Commission, 2011a), any reduction in the consumption

164 CPI into EU energy performance of buildings policy of energy is beneficial for achieving a reduction of GHG emissions. Rather than accepting the assumption that EU policy on energy efficiency indeed represents a case of high levels of CPI, I assess, in this chapter, the extent to which one energy efficiency policy measure in the EU – namely the energy performance of buildings policy – actually demonstrates CPI in its policy outputs and processes between 2000 and 2010.

5.1.1 Energy performance of buildings in the EU In 2010, the Buildings Performance Institute Europe (BPIE) carried out the first comprehensive study of the state of the EU’s building stock (BPIE, 2011). It is clear from this study that the EU’s building stock is not homogeneous: different member states have buildings of different types, age and energy performance. Some of the overarching statistics for the EU are nonetheless telling of the enormous challenge for EU policy to improve the energy performance of buildings.

While 25 per cent of the total of the EU’s building stock is non-residential (offices, retail, wholesale, public buildings, etc.), the remaining 75 per cent residential buildings are divided between owner-occupied and rented residences, with great variations among member states.82 In 2009, 42 per cent of the EU27 population lived in apartments, 34 per cent in detached houses and 23 per cent in semi-detached or terraced houses (BPIE, 2011). Additionally, the energy consumption of the building sector has increased steadily since 1990 (by about 1 per cent per year; European Commission, 2013b, p. 6), although the increase in energy consumption was greatest in non-residential buildings. Electricity consumption in non-residential buildings increased by 74 per cent between 1990 and 2010 (BPIE, 2011, p. 10). In general, trends in energy consumption in buildings are marked by a decrease in the consumption of solid fuels and oil and an increase in the use of natural gas and electricity and in overall energy consumption (European Commission, 2013b, p. 6). In Figure 5.3, we can see the evolution of housing space heating consumption compared to the growth in floor area in the EU27 between 1990 and 2008. It is evident that there have been efficiency improvements in that time period: although the total floor space has continued to increase, the total energy consumption for space heating has

82 Romania, for example, has a very high proportion of owner-occupied dwellings (at about 95 per cent), while the Czech Republic’s owner-occupied dwellings are about 55 per cent (BPIE, 2011, p. 11).

165 CPI into EU energy performance of buildings policy remained around the same levels. These absolute gains, however, are not yet reflected in reduced energy consumption.

Figure 5.3 Evolution in housing floor area and energy consumption for space heating in the EU27, 1990-2008.

Source: EEA, 2012, p. 33.

Reducing the energy consumption of buildings involves targeting the main sources of energy consumption, namely space heating (and cooling), water heating and the use of electric and electronic appliances (EEA, 2012, p. 32). Therefore, energy performance standards for buildings include measures for making buildings more secure against outside cold or hot air, through insulation in the building envelope (walls, roof, windows), combined with the latest airing technologies (ibid., p. 36). As energy performance standards gradually became more ambitious in the various member states, the general efficiency of new buildings has increased. One major challenge, however, remains the improvement of the energy performance of the existing building stock.

166 CPI into EU energy performance of buildings policy

According to BPIE’s report, about 40 per cent of the EU27’s building stock was built before 1960, when energy regulations were very limited. Less than 20 per cent of EU buildings were built between 1990 and 2010 (BPIE, 2011, pp. 9-10). This is crucial when considering the potential for advancing improvements in the energy performance of buildings in the EU – policies need to tackle measures in the existing building stock to impact the energy consumption of buildings. Thus, it is important also to take note of the rates of building renovations. As of 2013, the renovation rates for buildings in the EU were about one to two per cent per year (European Commission, 2013b). The type of renovations being undertaken is also crucial for effective reduction in energy consumption in buildings. BPIE proposes that renovations leading to a “nearly-zero energy building” hold the most promise for reducing energy consumption of buildings by 2050. This type of renovation includes (according to BPIE):

“the wholesale replacement or upgrade of all elements which have a bearing on energy use, as well as the installation of renewable energy technologies in order to reduce energy consumption and carbon emission levels to close to zero, or, in the case of an ‘energy positive’ building, to less than zero (i.e. a building that produces more energy from renewable sources than it consumes over an annual cycle). The reduction of the energy needs towards very low energy levels (i.e. passive house standards, below 15kWh/m2 [kilo Watt hours per metre squared] and year) will lead to the avoidance of a traditional heating system…” (BPIE, 2011, p. 103).

With such a type of renovation, BPIE argues that to have any chance of meeting 2050 targets for buildings, the entire building stock (100 per cent) of the EU27 will need to be renovated between 2010 and 2050 (a renovation rate of 2.5 per cent per year, with renovations meeting the nearly-zero energy building standard) (ibid., p. 109). These are important figures to note for measuring the level of CPI in the policy process and output of the EU’s EPB policy (see sections 5.3.1.1 and 5.3.2.2 below).

5.1.2 Energy performance of buildings policymaking

The less (fossil fuel-generated) energy consumed in buildings, the less GHG emissions from energy production and use. Analyses of how the EU will achieve its long-term climate policy objective of reducing GHG emissions in the EU by between

167 CPI into EU energy performance of buildings policy

80 and 95 per cent by 2050 all point to the substantial role to be played by energy efficiency measures (CANEurope, 2011; ECF, 2010; European Commission, 2011a; Heaps, Erickson, Kartha, & Kemp-Benedict, 2009; IEA, 2011a; Wesselink, Harmsen, & Eichhammer, 2010). Some estimate that energy efficiency measures can achieve about half of the 80-95 per cent emissions reduction goal by 2050 (Wesselink, et al., 2010). With buildings accounting for approximately 40 per cent of the EU’s total energy consumption (Andaloro, et al., 2010; Boasson & Wettestad, 2013; European Commission, 1984, 1998), measures to improve the efficiency of the economy must include energy performance measures for the buildings sector. Policy to improve the energy performance of buildings implies putting in place measures to tackle the energy waste through the building envelope and through the use of energy-efficient electric and electronic appliances and equipment within the building. Of the energy efficiency measures mentioned in Table 5.1 several of them already target energy efficiency measures for appliances, including, for example, the ecodesign Directive (2009/125/EC) and the energy labelling Directive (2010/30/EU). Although these pieces of legislation have an effect on the overall energy consumption of buildings,83 I focus here on the legislation to deal with the building envelope (windows, roofs, walls) and installed equipment (such as hot water boilers), namely the 2002 energy performance of buildings Directive and its 2010 recast Directive.

Several analysts have acknowledged that the complexity of improving the energy performance of buildings – due to the climatic variances across the EU, the ensuing differences in building standards, and the fact that the building sector is largely dominated by small businesses operating at a local level – represents a major challenge to the agreement on and implementation of energy performance measures in buildings. These barriers are linked to what is considered a “market failure” in terms of the promotion of energy efficiency (Dawson & Spannagle, 2009; Richter, 2010; Roosa, 2010). Among the difficulties related to the uptake in energy performance standards in the buildings sector are:

• Lack of awareness among consumers of the benefits of investing in measures to improve the energy performance of their homes;

83 The EEA reported in 2012 that these energy efficiency improvements “have not been able to keep up with increasing ownership and use of appliances. As a result, overall electricity consumption per dwelling for lighting and appliances has gone up, illustrating the rebound effect” (EEA, 2012, p. 33).

168 CPI into EU energy performance of buildings policy

• Split incentives, where a building constructor or owner may not be the final consumer of energy, and therefore has no incentive to ensure higher (and sometimes more expensive) energy performance standards in the building. This split incentive also plays a role where a landlord has little incentive to improve the energy performance of a building during renovation; • The initial purchase and construction costs for higher energy performance standards can be expensive, which can act as a deterrent; and, • Lack of knowledge of the various energy efficient technologies available, which may lead to a lock-in into older, less efficient technologies (Dawson & Spannagle, 2009, pp. 147-148; European Commission, 2011e, p. 6).

Additionally, there are certain barriers to introducing EU-level policy on the energy performance of buildings, such as subsidiarity rules and the lack of a clear advantage to making policy on the EU level on this issue, and the disparate construction industry that is generally made up of small to medium businesses operating in a locally- defined geographic area (without organised representation at the EU level; Boasson & Wettestad, 2013).

Possible policy measures can include minimum energy performance standards in appliances and buildings, energy efficiency targets, energy taxes, grants to promote energy efficient investments, among others. Home energy use has continued to increase over the last number of decades, even faster than population growth (Heaps, et al., 2009, p. 14), and this trend is expected to continue (EEA, 2012, p. 35). Such an increase is due to the increased income level in the EU, and the fact that most households have now adopted energy intensive appliances such as refrigerators, washing machines, dishwashers, tumble dryers and large televisions (EEA, 2012, p. 33). The services sector has seen rather more improvements in energy efficiency in terms of standard offices becoming more energy efficient since around the 1990s, although many of these efficiency gains have been offset by the rebound effect (Heaps, et al., 2009, p. 14; EEA, 2012, p. 22, 33). Thus, with business-as-usual (BAU) projections outlining an increase in the number of electronics and the size of homes, measures to improve the energy performance of buildings and appliances will need to continue to evolve to offset the increase in the use of heating and cooling and

169 CPI into EU energy performance of buildings policy appliances. Crucially, future policy will need to include measures on increasing the rate, and improving the type, of renovations in the existing building stock.

5.2 Measuring CPI in the EU’s EPB policy In this chapter, I measure and analyse the level of CPI in the policy output and process of the 2002 EPBD and the 2010 EPBD recast. To do so, it is necessary first to describe the benchmark against which I measure the levels of CPI.

5.2.1 Benchmark for measuring CPI in the policy process As outlined in chapter three, in which I described the conceptualisation and operationalisation of CPI, measuring the level of CPI in the policy process does not involve a single benchmark for a very high or complete level of CPI in the process. Rather, as outlined in Table 3.1, and based on a review of literature on EPI especially, it is possible to highlight a number of indicators for measuring the level of CPI in the policy process. In this study, I measure the level of CPI in the policy process with reference to the role of internal pro-climate stakeholders (namely, those policymakers within the EU institutions that hold a pro-climate stance, such as DG Environment, the environment committee (ENVI) of the Parliament, and the environmental Council formation); the involvement of external pro-climate stakeholders (stakeholders outside the EU policymaking institutions but who attempt to influence the policy process through consultation and lobbying and who have a pro-climate stance, such as environmental NGOs and/or energy efficient industries); and whether policymakers recognise (and articulate) the functional interrelations between long-term climate policy objectives and (in this case) energy performance of buildings policy during the process.

In the case of energy performance of buildings policy developed between 2000 and 2010, very high levels of CPI in the policy process would therefore see DG Environment as lead drafters for the policy proposals, with the ENVI committee in the Parliament drafting the reading reports for the Parliament’s plenary, and the environment Council drafting the Council’s position on the policy proposal. External pro-climate stakeholders would have easy access to provide input to the policy process and their pro-climate arguments and suggestions for amendments would receive general backing. Additionally, the motivation of policymakers to advance the energy performance of buildings policy would be to achieve long-term climate policy

170 CPI into EU energy performance of buildings policy objectives – in this respect, the recognition of the functional interrelations between the policy areas in the process would help push for more ambitious policy measures to improve the energy performance of the EU’s building stock by 2050 (given that reductions in energy consumption in buildings can directly help reduce GHG emissions).

5.2.2 Benchmark for measuring CPI in the policy output Establishing the level of CPI in the policy output of EU policy on the energy performance of buildings involves describing a benchmark for expected very high or complete levels of CPI in the final agreed policy. The distance between the actual policy output and this complete or very high level, and how much the output closes the gap between BAU and the very high or complete level of CPI, shows the level of CPI in the policy output. I first identify what sort of policy output for the energy performance of buildings would be in harmony with the 2050 goal of reducing GHG emissions in the EU by between 80 and 95 per cent compared to 1990 levels.

In the EPB sector, very high or complete CPI would imply significant improvements in the energy performance of the EU’s entire building stock. This can be measured in terms of reductions in the energy consumption of buildings. Very high or complete CPI in the energy performance of buildings policy output would then lead to decreases in the energy consumption of the entire EU building stock by 2050 in line with targets for decarbonisation. Importantly, such a benchmark implies that policy outputs must deal simultaneously with putting in place energy performance standards for new buildings and for existing buildings through renovations.

There have been several studies and scenarios describing the road to a decarbonised economy by 2050 in order to achieve 80-95 per cent reduction in GHG emissions in the EU. Regardless of whether the scenario focuses on a global or an EU-wide energy transition, all scenarios point to the great need, and potential, to improve energy efficiency measures to meet the long-term goals for GHG emission reduction (see ECF, 2010; EREC, 2010; EREC & Greenpeace, 2010; European Commission, 2011a; EWEA, 2011; ExxonMobil, 2012; Heaps, et al., 2009; Shell International BV, 2008; WWF, 2011). Few scenarios outline studies on the contribution of the buildings sector in particular, but several of these scenarios call for a realisation of the potential of retrofitting the current building stock at a faster pace and to high energy performance

171 CPI into EU energy performance of buildings policy standards, and of ensuring new builds quickly conform to high energy performance building standards, such as the passive building, zero-energy or energy-plus buildings (BPIE, 2011; EREC, 2010; Heaps, et al., 2009).84

In a system with 100 per cent of the EU’s building stock reaching very high standards of energy performance, it may be possible to conceive of reducing overall energy consumption in buildings to zero. However, in no scenario to 2050 is this considered a feasible option. This is for two main reasons. First, the rate of new builds and retrofitting of the existing building stock in countries with good levels of economic growth is about three per cent annually (Ries, Jenkins, & Wise, 2009, p. x). Rates of renovation or retrofitting in the EU27 in 2013 were assumed to be between one and two per cent annually (BPIE, 2011, p. 103; European Commission, 2013b, p. 7). This rate of retrofitting will have to be increased dramatically, and the type of renovations would have to ensure a very high standard of energy performance in buildings by 2050 to reduce energy consumption in buildings to as close to zero as possible. Second, there are several historical buildings and monuments that simply cannot be retrofitted to meet high-energy performance standards of buildings. Therefore, reasonable estimates show that the buildings sector can reduce its energy consumption to a maximum level of 200 Mtoe in the EU in 2050 (against c. 460 Mtoe of energy consumption in buildings in 2010; European Commission, 2011g) with a retrofitting rate of between 2.5 and 5 per cent, and achieving energy efficiency standards (e.g. a mix of “nearly zero energy buildings” and “energy plus” buildings) in about 90 per cent of the EU’s building stock (BPIE, 2011; Heaps, et al., 2009). With 200 Mtoe of energy consumption in the buildings sector representing a maximum level of consumption, a reasonable ideal level of CPI could expect to see a range of levels of energy consumption in buildings of between 100 (or less) and 200 Mtoe. Such a range allows for the continued higher levels of consumption in historical buildings, but assumes an aggressive push to retrofit the existing building stock so that it, and all new buildings, meet high energy performance standards. Figure 5.4 outlines what such a decrease in energy consumption (from 2000) looks like, and compares this very high level of CPI with the European Commission’s (2011) BAU scenario to

84 The passive house standard requires buildings to consume less than 15 kilowatt hours (kWh) of energy per m2 and per year, while energy plus buildings produce more energy through installed renewable energy generation sources than they consume (BPIE, 2011, p. 103).

172 CPI into EU energy performance of buildings policy

2050, which shows energy consumption of buildings to stand at 472 Mtoe in 2050. It is clear from the graph that very high levels of CPI are dramatically lower than BAU scenarios, so policy measures (discussed in the next section) need to close a significant gap.

Figure 5.4: Very high/complete CPI trajectory for energy consumption in buildings in EU27, from 2000 to 2050, compared to the BAU trajectory (that includes actual levels of consumption from 2000 to 2010; measured in Mtoe).

Source: European Commission, 2011g; Heaps, et al., 2009.

In the discussion on the 2002 EPBD and its 2010 recast below, I will be able to assess the distance between the policy output of these policies and the very high/complete trajectory for CPI of achieving a reduction in energy consumption in buildings to 100 to 200 Mtoe by 2050.

5.3 EU EPB policy Many of the early pieces of energy efficiency legislation described above (see Table 5.1) were relevant for improving the energy performance of buildings. The 1989 construction products Directive aims to promote the free movement of construction products, with provisions setting out criteria for the health and safety of the materials, and highlighting the need for ensuring the products promote energy efficiency. Annex one of the Directive explained that “construction works and its heating, cooling and

173 CPI into EU energy performance of buildings policy ventilation installations must be designed and built in such a way that the amount of energy required in use shall be low…” (Directive 89/106/EEC). Next, the 1992 labelling Directive focused specifically on the provision of product information on the consumption of energy and other resources by household appliances (Directive 92/75/EEC). This Directive aimed to ensure consumers had sufficient information on the energy use of such household products at the time of purchase. It applied to refrigerators, freezers, washing machines, tumble dryers, dishwashers, ovens, water heaters and hot-water storage appliances, lighting sources, and air-conditioning appliances. The 1992 hot-water boilers Directive (92/42/EEC) determines the essential efficiency requirements for hot-water boilers, which then receive the “CE” marker indicating their conformity with (all applicable) legislation. The SAVE (Specific Actions for Vigorous Energy Efficiency) programme of the early 1990s aimed to provide financial support for actions to improve energy efficiency, and to provide specific targets for member states (see Council Decision 91/565/EEC). The later 1993 SAVE Directive was an overarching Directive aimed at limiting carbon dioxide emissions by improving energy efficiency (an explicit attempt to make energy efficiency measures a direct response to the problem of climate change), notably through programmes in the fields of energy certification of buildings; thermal insulation of new buildings; regular inspection of boilers; among others (Art. 1), but it was already limited in scope by the Council, in that it did not include specific commitments for member states. With hindsight, the success of the SAVE programme (and many later energy efficiency measures also) was hampered by the lack of member state ambition and the limited financing available to implement its objectives (Boasson & Wettestad, 2013; Oberthür & Roche Kelly, 2008).

Against the background of failure to achieve energy efficiency targets and weak policy measures, it became clear that further legislation would be required to spur improvements in energy efficiency (Henningsen, 2011). The 2002 EPBD and its 2010 recast represent the main pieces of legislation on buildings under examination in this chapter. The aims of these pieces of legislation are to improve the energy performance of buildings by requiring member states to adopt technological measures to improve the quality of buildings. The main elements of the Directives include: a method for calculating the energy performance of buildings; requirements for national building codes; requirements on the control of heating and cooling systems in

174 CPI into EU energy performance of buildings policy buildings; recommendations for financial support measures; and outlines for national energy certification of buildings (Boasson & Wettestad, 2013, p. 130; Directives 2002/91/EC and 2010//31/EU).

5.3.1 Energy performance of buildings Directive (2002/91/EC) Both the 1984 and 1998 strategy documents published by the Commission set the frame for future policy proposals on improving the energy performance of buildings (European Commission, 1984, 1998). Citing that buildings account for 40 per cent of the EU’s energy consumption, the Commission targeted the energy performance of buildings as one area with high potential for improvement (European Commission, 1998, p. 11). The Council backed Commission calls for more ambitious policy development in a Resolution on energy efficiency on 7 December 1998 (98/C 394/01), in which it reaffirmed the “importance of further development and implementation of appropriate common and coordinated policies and measures (CCPMs) for the energy efficiency sector” (paragraph 3). The Commission then set about drafting a proposal for a Directive on the energy performance of buildings (Boasson & Wettestad, 2013; European Commission, 2000a).

The legislative proposal, drafted by DG Transport and Energy (DG TREN, or “DG Energy” today, and hereafter) under the leadership of Commissioner Loyola de Palacio, was published in May 2001 (European Commission, 2001d). This proposal was made with Article 175 of the Treaty establishing the European Community as the legal basis (Amsterdam Treaty, TEC: today’s Article 192 of the Treaty on the Functioning of the EU, TFEU), placing the proposal within the framework of the environmental chapter of the Treaty. 85 Again, climate change and the EU’s commitments to reduce its GHG emissions are cited in the proposal as a major rationale for taking action at EU-level on improving the energy performance of buildings (European Commission, 2001d, pp. 15, 19). Such a strategy follows along with much of the EU’s energy policy proposed in the 2000s, linking the energy issue to either internal market or environmental competences to push legislation forward (see also chapter two). The overall objective of the proposal was to set minimum standards, and suggested the following requirements:

85 See Annex II for the text of Article 175, TEC.

175 CPI into EU energy performance of buildings policy

a) A general framework for a common methodology to calculate the energy performance of buildings; b) The application of minimum energy performance standards in new buildings; c) The application of minimum energy performance standards in large existing buildings that are to undergo renovation; d) The energy certification of buildings, and the display of this energy information in public buildings, and; e) The regular inspection of boilers, air-conditioning and heating systems in existing buildings (European Commission, 2001d, pp. 21-22).

The proposal moved into the first reading stage in the Parliament on 17 May 2001, after a first debate in Council on 14 May (Council of the European Union, 2001a). In the Parliament, the industry, external trade, research and energy committee (ITRE), with Spanish MEP Aledo Vidal-Quadras of the European People’s Party (EPP) group as rapporteur, was responsible for drafting the first reading opinion. Cristina Garcia- Orcoyen Tormo, also a member of the EPP group, acted as rapporteur for providing the opinion of the environment, public health and consumer policy committee (ENVI). The first reading committee report was brought to the Parliament’s plenary on 18 December 2001, with debate in the plenary taking place on 4 February the following year, and the adoption of the opinion on 6 February 2002.

The first reading opinion of the Parliament on this file did not deviate too radically from the original proposals of the Commission. It did, however, already indicate some hesitation with regard to proposals on the renovations of existing buildings. The Parliament underlined the need for financial mechanisms to ensure such energy efficiency measures can be implemented in renovations of existing buildings (European Parliament, 2002a, p. 2). The cost-effectiveness of the proposed measures, and the role of member states in incentivising the uptake of energy efficiency measures were emphasised in the opinion: thus the issue of financing was one of the main points for the Parliament in the ensuing co-decision negotiations. In addition, the Parliament proposed new articles covering evaluation and information measures, to ensure follow-up of member state implementation and consumer awareness (ibid., pp. 11-12).

176 CPI into EU energy performance of buildings policy

The Commission accepted several of the 30 amendments proposed by Parliament and published an amended proposal on 16 April 2002. The emphasis of the Parliament on cost-effectiveness was accepted in the Commission’s new proposal, as were some of the Parliament’s suggestions for member state support of energy efficiency measures (for example, tax deductions, low interest credits, funding through public programmes; European Commission, 2002a, pp. 7, 8). The Council position on this amended proposal followed on 7 June that same year.

Warnings had come earlier in the policy process that the Council would be less interested in stringent measures on the energy performance of buildings. Reports from an informal Council agreement in December 2001 suggested Council wanted to delay the implementation timetable of the measures in the proposed Directive by up to five years beyond the suggestion of the Commission. It also aimed to water down the requirements for existing building renovations, qualifying the application of energy efficiency measures to “major” renovations only (ENDS Europe, 2001d). It was therefore no surprise that the Council’s position, published on 7 June 2002, included further flexibility for member states in applying the minimum standards for energy performance, and added the proviso that standards will apply in existing buildings only undergoing “major renovation” – a phrase left to the interpretive discretion of the member states themselves. The time period for implementing the provisions of the Directive was finally set by Council at 36 months, with additional flexibility so that member states could benefit from up to four additional years before fully implementing the provisions of the Directive (Council of the European Union, 2002, pp. 13, 14).

Not being able to agree on the deadline for the implementation of the proposed Directive, the Parliament began its second reading of the draft. The committee second reading recommendation was tabled for plenary on 11 September 2002, and the final second reading opinion of the Parliament was adopted on 10 October. News reports declared a “soft” attitude of the Parliament towards the Council’s amendments, voting in committee in favour of a six-year timetable for the implementation of the provisions of the Directive (ENDS Europe, 2002), rather than the quicker implementation it called for in its first reading. The second reading recommendation text stated: “member states may, because of lack of qualified and/or accredited experts, have an additional period of three years to apply fully the provisions…”,

177 CPI into EU energy performance of buildings policy where the Council position called for an additional four years (European Parliament, 2002b). The Commission and Council adopted the amendments proposed by Parliament in its second reading and the final text was signed on 16 December 2002.

Since then, there has been much criticism of the Directive, particularly in terms of the lacking level of and slow implementation of its provisions that followed. In fact, commentators describe the Directive as having “little impact” on the energy performance of buildings overall in the EU, due in part to its limited scope, and in part to the great flexibility left to member states to implement the Directive as they wish and when they wished (Henningsen, 2011, p. 133). Some stakeholders, including the insulation and glass industry associations, and environmental NGOs, were already calling for further legislation on the issue in 2004, citing insufficient action, poor implementation and difficulties to capitalise on the potential to reduce GHG emissions through the policy measures under the energy performance of buildings (ENDS Europe, 2004, 2005). These calls resulted in a proposal for a recast of the Directive in 2008.

5.3.1.1 CPI in the policy process of the 2002 EPBD As outlined in chapter three (see Table 3.1), measuring CPI in the policy process involves examining three main factors: the involvement of internal pro-climate stakeholders; the involvement of external pro-climate stakeholders; and, the recognition in the policy discourse of the functional interrelations between energy performance of buildings policy (in this case) and long-term climate policy objectives (see also Table 5.2).

Internal pro-climate stakeholders Internal pro-climate stakeholders in the policy process leading to the 2002 EPBD only had a low level of involvement. In the Commission, DG Energy drafted the proposal for the 2002 EPBD. Although there is no indication of a special role for DG Environment in the drafting process, internal procedures within the Commission allowed space for this DG to comment on the proposal. When the proposal came to the college of commissioners, three commissioners objected to the need for a Directive on the issue based on subsidiarity grounds: Chris Patten, then commissioner for external relations, Neil Kinnock, then Vice President of the Commission, and Frits Bolkestein, then commissioner for the internal market and services. Although they

178 CPI into EU energy performance of buildings policy were overruled, their objection led to further time redrafting the proposal before its publication. The redrafting was again in the lead hands of DG Energy, under commissioner Loyola de Palacio. Another point of note in the development of the proposal in DG Energy is that it was rather DG Enterprise that pushed for certain measures in the proposal, such as the importance of certifying buildings that have over 1000m2 of floor space: a conclusion that came out of a working group on sustainable construction set up with industry members and member state representatives by DG Enterprise in 2001 (Boasson & Wettestad, 2013, p. 135). Thus, DG Environment did not seem to play an active role in pushing for more ambitious measures than were already outlined by DG Energy. This implies a rather low level of involvement of internal pro-climate stakeholders in the Commission, with DG Environment able to provide opinions to the process if it wished, but not playing a major role in the internal consultations in the Commission (see Table 3.1).

In the Parliament, although the ENVI Committee provided an opinion to the ITRE committee, it did not seem to influence the final weakened Directive. With the main amendments in the Parliament related to funding, cost-effectiveness, and implementation deadlines (which were finally agreed in opposition to strong climate objectives), the procedural option of providing an opinion did not lead to substantive influence from the ENVI committee. The ENVI committee’s opinion suggested amendments that were not taken on board. One amendment that may have had a substantial impact on the ability of the EPBD to achieve its expected policy output was to specify that all buildings with a floor area of over 500m2 (rather than 1000m2 as proposed by the Commission and as adopted in the final Directive) undergoing renovations would be required to meet minimum energy performance standards (European Parliament, 2001b, p. 31). Thus, the involvement of pro-climate stakeholders in the Parliament is low to medium, where the ENVI committee provided an opinion, but this opinion was not necessarily taken on board or reflected in the process.

In the Council, the discussions on the legislative proposal took place within the Council of energy ministers. The proposal was first discussed in the industry and energy Council meeting in December 2001, where already the general outlines of the Council’s position to promote further member state flexibility in the final Directive were agreed (Council of the European Union, 2001a; ENDS Europe, 2001d). The next

179 CPI into EU energy performance of buildings policy occasion when the issue was discussed led to the adoption of the common position, again in the industry and energy Council meeting in May 2002 (Council of the European Union, 2002). The Council approval of the final text occurred in the transport, telecommunications and energy council meeting in November 2002. As is clear throughout the process, the environment Council was not involved in the development of the Council’s position. This leads to a very low level of involvement of internal pro-climate stakeholders in the policy process in the Council.

Overall, and in aggregate, a low level of involvement of internal pro-climate stakeholders in the EPBD policy process is evident (see Table 3.1). While some opportunities existed for pro-climate stakeholders to be involved in some of the co- deciding institutions, either they could not or did not take advantage of these opportunities (e.g. in the Commission) or their opinions were disregarded in the policy process (e.g. in the Parliament). In some respects, as combating climate change was one of the stated motivations behind the EPBD proposal (along with strong statements on improving energy security), it may be that internal pro-climate stakeholders felt their colleagues could promote climate policy objectives without their explicit involvement. As is clear from the poor final performance of the EPBD through implementation (see above), this did not help climate policy move forward through the EPBD.

External pro-climate stakeholders The level of involvement of external pro-climate stakeholders in the policy process leading to the 2002 EPBD is low. First, it is clear that environmental NGOs working at the EU-level in the early 2000s did not focus on energy efficiency measures. In fact, interviewees have indicated that none of the major environmental NGOs working at EU-level had staff dedicated to working on policy advocacy to improve energy efficiency at this time (interviews 5, 6, 10 & 11). Therefore, the NGOs rather responded to developments as they occurred than trying to spur action at EU-level. The task fell to already overburdened staff within the NGOs to follow the developments on this dossier (interviews 5 & 6), and often the energy efficiency policy development was neglected by NGOs. These staff members were able to respond only in a limited way and only when major developments occurred in the institutions. Greenpeace, for example, came out in support of the Parliament in its first reading, when it seemed the Parliament would resist Council pressures to extend

180 CPI into EU energy performance of buildings policy the implementation deadline of the Directive (ENDS Europe, 2001b). In this case, although opportunities existed for informal lobbying during the policymaking process, environmental NGOs rather lacked the expertise and the resources at this time to push policymakers to make strong pro-climate policy.

Second, in terms of other organisations beyond environmental NGOs that could have pushed for ambitious policy, there were three industry associations that were more active and did provide some input to policymakers. These included EuroACE (the European Alliance of Companies for Energy Efficiency in Buildings), FIEC (the European Construction Industry Federation), and Eurima (the European Insulation Manufacturers Association). Each of these associations lobbied the EU institutions to a certain extent to enact strong legislation on the energy performance of buildings, underlining especially job-creation opportunities in moving forward on energy efficiency, along with climate benefits (ENDS Europe, 2001d). Even with this limited level of lobbying, FIEC, EuroACE and Eurima were more “reactive” than “proactive” in the policy process leading to the 2002 EPBD (Boasson & Wettestad, 2013, p. 135; interview 10). As one interviewee described it, at this time, his organisation was not very effective at lobbying the EU institutions on the EPBD in the early 2000s, saying they “never quite dotted the i’s and crossed the t’s on that first Directive” (interview 10). Although these associations reacted against warnings that the Council would extend the implementation timetable, they did not effect great change in the final agreement. Many of these associations would go on to call for a tightening of the legislation on energy performance of buildings not long after it was agreed (ENDS Europe, 2004, 2005).

Third, there were few, if any, official procedures and avenues for stakeholders to access policymakers beyond their own informal lobbying activities at this time. There was no systematic public, open consultation in advance of the Commission’s proposal. This legislative proposal was made in 2001, thus before regulatory impact assessments and public consultations were introduced in the Commission (in 2002),86 so the Commission did not carry out an impact assessment. Neither did the Commission’s proposal indicate any formal or informal consultation with external actors. It is clear that some of the Commission’s proposal included points raised in

86 See: http://ec.europa.eu/governance/better_regulation/impact_en.htm, accessed: 16 October 2012.

181 CPI into EU energy performance of buildings policy working groups of the early European Climate Change Programme (ECCP) process,87 which aimed to involve member state, industry and NGO representatives in discussions on potential policy measures to combat climate change, yet this process was rather opaque. It is unclear which organisations were involved in pushing policy in the energy performance of buildings particularly. However, industry members and member states (who may not necessarily promote climate policy objectives) certainly had an extra opportunity to express their interests in the working group on sustainable construction set up in 2001 (Boasson & Wettestad, 2013). In this working group, environmental NGOs had no role.

In terms of assessing the level of external climate policy stakeholder involvement in the policy process, it can be considered low (see Table 3.1). There was no open consultation process in advance of the Commission’s proposal, although it does appear that the energy performance of buildings issue may have been discussed under the ECCP (European Commission, 2001d). Pro-climate stakeholders, whether from environmental NGOs or from industries interested in pushing for ambitious policy, did not play much of a role in the policy process – there was no or little access to the policy process, no guaranteed procedure allowing access, and their opinions did not play a major role in the process. The peculiarity of this time period is the lack of dedicated expertise or resources among environmental NGOs on energy efficiency issues, meaning that it was the industry associations working on energy efficiency measures that played a more substantial (yet still limited) role in the process. These industry associations were more likely to use job-creation arguments in their lobbying than climate policy arguments. However, even the industrial actors did not play a major role in the policy process in the lead up to the 2002 EPBD (interviews 5, 6, 10, 11 & 12).

Recognition of functional interrelations The extent of the recognition of functional interrelations between climate policy objectives and the EPBD objectives is medium for the 2002 EPBD policy process. From the very outset of the policy process, policy on the energy performance of buildings was linked to combating climate change and achieving the objective of

87 The ECCP was established in 2000 and was a multi-stakeholder consultative process that aimed to discuss potential policy measures to combat climate change (European Commission, 2000a). See also chapter two.

182 CPI into EU energy performance of buildings policy reducing GHG emissions in line with the EU’s commitment under the Kyoto Protocol. In its proposal, the Commission mentioned that concerns about climate change, and concerns about the rising dependence of the EU on energy imports, were reasons for economising the “use of energy wherever possible” (European Commission, 2001d, p. 2). The Commission also placed climate policy objectives in the early recitals of the proposed Directive, with Recital 3 reading as follows: “increased energy efficiency constitutes an important part of the package of policies and measures needed to comply with the Kyoto Protocol…” (European Commission, 2001d, p. 19). This recital was not amended by the Parliament or the Council in the policy negotiations. The functional interrelations between reducing energy consumption in buildings and combating climate change by reducing energy-related GHG emissions were clearly recognised, at least at the outset of the process.

When it comes to the long-term climate policy objectives of reducing GHG emissions in the EU by 80-95 per cent by 2050 (or ensuring that global temperature increase is limited to 2°C compared to pre-industrial levels), neither the proposal drafted by the Commission, nor the amendments by Parliament and Council, nor the final Directive itself can be considered as ambitious enough. The Directive does not sufficiently emphasise improvements in the existing building stock, which is crucial for the achievement of significant reductions in time for 2050. Indeed, the EU’s building stock includes 40 per cent of buildings that were built before 1960, and many buildings that were built in 1910 or earlier (BPIE, 2011), when energy performance standards were non-existant. Additionally, the Directive outlined that it was up to member states to decide on and implement the energy performance standards for new buildings (Recitals 9, 10, 12; Arts. 3, 4). The EU institutions did not translate the recognition of the functional interrelations between the energy performance of buildings policy and climate policy objectives into ambitious policy measures. Neither did they consider long-term climate policy measures beyond the Kyoto Protocol commitments (to reduce GHG emissions in the EU15 by eight per cent by 2008-2012 compared to 1990 levels). In line with the indicators outlined in Table 3.1 a medium level of recognition of the functional interrelations is evident in this case. Functional interrelations were recognised at the beginning of the policy process, but this recognition did not motivate more ambitious policy measures.

183 CPI into EU energy performance of buildings policy

5.3.1.2 CPI in the policy output of the 2002 EPBD Energy consumption in buildings stood at 447 Mtoe in 2000. With energy consumption of the EU’s buildings (EU27) rising from 423 Mtoe in 1990 to 447 Mtoe in 2000, continuing such a trend would lead us to expect energy consumption of buildings to hit about 471 Mtoe in 2010 without any policy measures, under a BAU scenario. In its proposal for the EPBD, the Commission stated that the potential for improvement in the energy performance of buildings could amount to an energy consumption reduction of 22 per cent between 2000 and 2010 (European Commission, 2001, p. 8). From the 447 Mtoe of energy consumption in buildings in 2000, a 22 per cent reduction would result in a reduction of about 98 Mtoe by 2010 (resulting in 2010 energy consumption in buildings of 349 Mtoe). This is regarded by the Commission as a feasible, but ambitious, reductions potential. The Commission goes on to state that the indicative energy savings target agreed in the Council’s resolution on 7 December 1998 on energy efficiency (98/C 394/01) to improve the energy intensity of final consumption in the EU would result in avoided energy consumption of about 55 Mtoe by 2010 (in other words, avoided consumption compared to what would otherwise have been consumed; ibid.). Avoiding 55 Mtoe of BAU expected levels of 471 Mtoe energy consumption in buildings in 2010, through measures of the EPBD, would result in 416 Mtoe energy consumption of buildings in 2010. Thus, EPBD policy measures were expected to result in energy consumption of buildings at 416 Mtoe by 2010.

184 CPI into EU energy performance of buildings policy

Figure 5.5: CPI in the policy output of the 2002 EPBD, with measures expected to avoid the consumption of 55 Mtoe of energy in buildings in the EU between 2000 and 2010 (compared to BAU and to very high CPI range outlined in Figure 5.4).

Source: European Commission, 2001d; own calculations.

Although these figures are part of the Commission’s proposal, the expected policy output of the final Directive is similar. These figures are not adjusted downwards in consideration of the long timetable for member states to implement the provisions of the EPBD and the non-binding or flexible nature of many of the measures of the Directive. Such a policy output means that these figures should be considered with some level of scepticism. Yet, for the sake of establishing the potential level of CPI in the policy output, I examine this against the expected output of the EPBD avoiding 55 Mtoe of energy consumption in buildings by 2010. This output would succeed in closing part of the gap between expected BAU and CPI scenarios to 2010 (see Figure 5.5). Energy consumption of buildings was expected to hit 471 Mtoe in 2010 without the EPBD. The range of scenarios for CPI for 2010 is between 377 and 397 Mtoe of energy consumption in buildings by 2010 (see also Figure 5.4). Thus, the policy output in the 2002 EPBD (of 416 Mtoe in 2010) closes the gap between BAU (471 Mtoe) and CPI ranges by between 59 and 74 per cent. As outlined in Table 3.2, this results in medium to high levels of CPI for the policy output of the 2002 EPBD.

This cannot be considered a realistic measurement for the level of CPI, however. First, implementation of the EPBD provisions was delayed in member states by between

185 CPI into EU energy performance of buildings policy three and seven years after its entry into force, delaying the impact of the provisions even further. Second, the expected impact of the EPBD is based on an indicative target (non-binding) that relied on member state implementation. Later analysis shows the inadequate implementation across the EU of the EPBD measures, and much literature describes the EU’s implementation deficit (see, e.g., Haverland & Romeijn, 2007; Lampinen & Uusikylä, 1998). Third, actual energy consumption of buildings between 2000 and 2010 rose significantly (from 447 Mtoe in 2000 to 482 Mtoe in 2010), even beyond the BAU scenarios suggested in Figure 5.5. Examining the implementation reality after the EPBD shows that the final policy outcome meant that the EPBD had either no effect on reducing energy consumption in buildings between 2000 and 2010, or, at best, resulted in the slight reduction (of two Mtoe) of energy consumption in buildings recorded between 2005 and 2010 (see Figure 5.6).88

Figure 5.6: EPBD measures compared to actual consumption of energy in buildings, 2000-2010 (measured in Mtoe).

Source: European Commission, 2001d; Eurostat data; own calculations.

Figure 5.6 shows also the actual consumption of energy in buildings between 2000 and 2010. Where an ideal CPI trajectory range shows the need for energy consumption to be reduced between 2000 and 2010 by between 50 and 70 Mtoe, actual energy consumption in the same time period increased by 36 Mtoe. This compares even more dramatically to the Commission’s original proposed potential for

88 Figures based on Eurostat data: http://ec.europa.eu/eurostat, accessed: 13 November 2012.

186 CPI into EU energy performance of buildings policy the EPBD to achieve a reduction in energy consumption in the buildings sector of 98 Mtoe between 2000 and 2010 (a potential reduction in consumption of 22 per cent by 2010 compared to 2001 levels; European Commission, 2001d, p. 8), or even the expected avoidance of 55 Mtoe in 2010 compared to BAU due to the EPBD (ibid.).

Energy consumption of buildings did fall by (just) 2 Mtoe between 2005 and 2010 (European Commission, 2011g). While energy consumption had risen between 1990 and 2005 by 62 Mtoe, without the implementation of the provisions of the EPBD, one could have expected energy consumption to continue rising to 2010 and beyond. Continuing such a rising trend (in a simplistic linear fashion), we could have expected to see on average a further 21 Mtoe of energy consumption in the buildings sector between 2005 and 2010, which would have resulted in energy demand in the buildings sector of 506 Mtoe by 2010. Instead, the actual energy consumption of buildings was recorded at 23 Mtoe lower in 2010 (483 Mtoe). However, while some of this reduction could potentially be attributed to a functioning EPBD, much of this avoided energy consumption can also be considered a direct effect of the financial and economic crises from 2008 onwards, placing doubt on the actual impact of the EPBD in this time period (Henningsen, 2011). While a reduction of 2 Mtoe between 2005 and 2010 (to 483 Mtoe) could be considered in a positive light, it remains difficult to assess the extent to which the EPBD played a role in the reduction of energy consumption in this time period. Even with the assumption that the EPBD was a major factor in the small reduction in energy consumption, the level in 2010 is still a considerable way from an ideal level of between 377 and 397 Mtoe. In fact, if BAU for 2010 hit 506 Mtoe, and the actual policy impact of the EPBD was the avoidance of 23 Mtoe, then the EPBD could have closed the gap between BAU and CPI ranges for energy consumption of buildings by just about 18 to 22 per cent. Even taking the EPBD output in a positive light, such a measurement would arrive at CPI in the policy outcome/impact of the EPBD of very low to low levels (see Table 3.2).

What this empirical evidence tells us is that although the policy output of the 2002 EPBD could have resulted in medium to high levels of CPI based on an assessment of the expected impact of the policy measure, this medium to high level is inaccurate and tempered by several factors. First, it is tempered by the fact that implementation of the Directive was expected only from 2006 – limiting the time frame for achieving the expected results to just 2006-2010, and thus reducing the likelihood that the expected

187 CPI into EU energy performance of buildings policy output would be achieved. Second, the energy efficiency targets agreed in Council in 1998, to which the EPBD contributes, are non-binding. The non-binding nature of the target may have further limited the implementation success at member state level. These are issues that will need to be kept in mind when discussing CPI in the policy output of the 2010 EPBD (see section 5.3.2 below), for which there is not yet sufficient empirical data on the evolution of the energy consumption in buildings (as of March 2013).

In conclusion, a reasonable assessment of the level of CPI in the policy output for the 2002 EPBD would lie somewhere between the expected results (medium to high levels of CPI) and the empirical reality of the EPBD’s impact (very low to low levels of CPI). A generous assessment of CPI in the policy output implies taking account of the tempering factors discussed above, while also giving the benefit of the doubt to the ambitions for the policy output of the EU in agreeing the Directive in the first place. However, it is impossible to ignore the time-delay for implementation of the EPBD provisions and the non-binding nature of the final policy output that made any achievement of the potential of the EPBD doubtful. Such a consideration places slightly more weight on the final very low to low levels of CPI. Thus, the level of CPI in the policy output of the 2002 EPBD could be described as low (see also Table 5.2).

5.3.2 The 2010 recast of the EPBD (2010/31/EC) With calls so soon after the entry into force of the 2002 EPBD for updated and new EU legislation on the energy performance of buildings (ENDS Europe, 2004, 2005), and studies following soon thereafter on the limitations, and the implementation failures, of the EPBD (Henningsen, 2011), the Commission published a proposal for a recast of the EPBD in 2008. The proposal came in the context of discussions to move to a broader energy efficiency target to combat climate change. The Commission’s 2005 green paper on energy efficiency suggested a target of cutting energy consumption by 20 per cent by 2020 compared to 2005 levels (European Commission, 2005b, p. 4). The European Council later endorsed an indicative non-binding target of saving 20 per cent of the “EU’s energy consumption compared to projections for 2020” by 2020 (European Council, 2007, p. 20). Then energy Commissioner Andris Piebalgs declared that energy efficiency would be a top priority during his term (Henningsen, 2011, p. 131). The priority for energy efficiency may not be so clear in

188 CPI into EU energy performance of buildings policy later negotiations on the 2020 targets in the climate and energy package (20 per cent reduction in GHG emissions, and 20 per cent increase in renewable energy by 2020), with the non-binding 20 per cent energy savings target somewhat falling into the background, as it did not feature in the concrete policy measures of the “climate and energy package” (see also chapter two; European Commission, 2008a; see, Oberthür & Pallemaerts, 2010b; Boasson & Wettestad, 2013).

The 2008 second strategic energy review outlined the objectives of negotiating a recast of the 2002 EPBD. The aim was to “extend the scope, simplify its implementation and develop energy performance of buildings certificates into a real market instrument” (European Commission, 2008d, p. 11). The proposal for a recast of the EPBD, which was released together with the second strategic energy review, kept the general framework of the first EPBD focusing on calculation methodologies; recommendations for financial support; certification measures; requirements for heating and cooling systems; and requirements for national building codes, but it did also extend the 2002 Directive. It proposed that all buildings undergoing major renovations would have to meet minimum energy performance standards (no longer with the 1000m2 threshold), that energy certificates would have to be used in all sales and rental advertisements, and that the inspection of boilers would have to be improved (European Commission, 2008c). The proposal also aimed to encourage member states to develop national policies for the greater uptake of low- and zero- energy buildings (ENDS Europe, 2008n; European Commission, 2008c, p. 6).

The Parliament took up the first reading of the proposal, again with the ITRE committee drafting the first reading opinion. Silvia-Adriana Ţicău, a Romanian MEP from the group of the Progressive Alliance of Socialists and Democrats (S&D), acted as rapporteur. Among the five shadow rapporteurs, MEP Claude Turmes represented the Greens.89 The first reading opinion, adopted by the Parliament on 23 April 2009, was already stronger than the Commission proposal in many respects. The rapporteur pushed for financial incentives to encourage energy efficiency measures, and MEP Fiona Hall, member of the alliance of liberals and democrats for Europe political group (ALDE) and shadow rapporteur, pushed for tightening proposals to include

89 The other four shadow rapporteurs were: Algirdas Saudargas of the EPP group; Fiona Hall of the ALDE group; Vicky Ford of the European Conservatives and Reformists group; and, Marisa Matias of the European United Left – Nordic Green Left group.

189 CPI into EU energy performance of buildings policy targets for all new buildings to be zero-energy by 2020 (Boasson & Wettestad, 2013; ENDS Europe, 2009e). The final first reading report, with 107 amendments to the Commission proposal, included both of these proposals, with a 2019 deadline suggested for zero-energy buildings, and a call for an energy efficiency fund to be created by 2014 (ENDS Europe, 2009f; European Parliament, 2009a, 2009c).

Member states in the Council were less enthusiastic about strengthening the Commission’s proposal for a recast, and there were early indications of reticence in applying minimum standards to all buildings undergoing major renovations – a major element in the Commission’s proposal (Boasson & Wettestad, 2013; ENDS Europe, 2009d). Council debated the proposal in its transport, telecommunications and energy formation on both 12 June 2009 and 7 December 2009, before publishing its position on 14 April 2010 (Council of the European Union, 2009a, 2009b, 2010). In the run up to the adoption of its agreed position, the Council complained that Parliament’s amendments were overly ambitious (ENDS Europe, 2009b), and informal negotiations with the Parliament (in trilogues, with the rapporteur of the ITRE committee) were ongoing until the adoption of its position (Council of the European Union, 2010, pp. 55-56; ENDS Europe, 2009g). The final position showed the fruits of these negotiations, in that a deadline for achieving “nearly zero-energy buildings” was set for 2020, with public buildings to achieve this goal in 2018. The definition of nearly zero-energy buildings refers to achieving a very highly-efficient energy performance,90 the result of a compromise between Parliament and Council (Council of the European Union, 2010; ENDS Europe, 2009a). However, it is up to the member states to decide on the standard meant by “nearly zero-energy”.

The Parliament responded very quickly with its second reading recommendation, published on 18 May 2010, approving the Council’s position. The final Directive (2010/31/EU) was published in the Official Journal on 18 June 2010.

90 “Highly-efficient” is described in paragraph one of Annex I of the 2010 EPBD as “determined on the basis of the calculated or actual energy that is consumed in order to meet the different needs associated with its typical use and shall reflect the heating energy needs and cooling energy needs (energy needed to avoid overheating) to maintain the envisaged temperature conditions of the building, and domestic hot water needs”.

190 CPI into EU energy performance of buildings policy

5.3.2.1 CPI in the policy process

Internal pro-climate stakeholders The involvement of internal pro-climate stakeholders in the policy process leading to the adoption of the 2010 EPBD recast is low. It was again DG Energy in the Commission, the ITRE committee in the Parliament, and the transport, telecommunications and energy Council that were in the lead in the negotiations. However, through the usual consultation procedures, there were opportunities for internal pro-climate stakeholders to at least be involved in the policy process.

In the Commission, DG Environment was one of eight DGs involved in the impact assessment process, but its overall involvement is unclear. Procedures were in place to allow DG Environment to play a role, yet many of the DGs of the Commission were in favour of more ambitious policy on the energy performance of buildings. It is plausible that DG Environment did not see the need to be greatly involved, especially with DG Energy pushing for ambitious policy (interview 12). This implies a very low to low level of involvement of internal pro-climate stakeholders in the Commission, as, considering the procedures allowed for consultation, the pro-climate actors were not visible in the process. These actors themselves did not seem to take advantage of such opportunities, but may not have seen the need to do so.

Within the Parliament, the evidence points to more interaction with pro-climate stakeholders, although not in terms of ENVI committee involvement. The rapporteur on the dossier, although a member of the progressive alliance of socialists and democrats, had a pro-climate stance, and became a member of the cross-party renewable energy association of parliamentarians, EUFORES. In addition, one of the shadow rapporteurs was Green MEP Claude Turmes, who was often outspoken on the need to improve the environmental ambition of the proposal and to promote energy efficiency as a top priority (ENDS Europe, 2009a, 2010c). MEP Fiona Hall, of ALDE, was also influential in the policy negotiations in pushing for more ambitious policy (see also Boasson & Wettestad, 2013). In this case, the level of involvement of pro- climate stakeholders in the Parliament is low to medium, with pro-climate shadow rapporteurs playing a role in the drafting of the Parliament’s reports and opinions, but the ENVI committee being rather absent.

191 CPI into EU energy performance of buildings policy

Within the Council, the main pro-climate stakeholder involved in the policymaking process can perhaps be identified as the Presidency at the time of the final negotiations – Sweden (in the second half of 2009). Although the environmental Council was not involved, the general green credentials of Sweden (Liefferink, Arts, Kamstra, & Ooijevaar, 2009) may have helped push the dossier through. One interviewee commented that the Swedish Presidency aimed to agree on the file before the Copenhagen climate summit in December 2009. Although the official agreement on the dossier did not come until the publication of the Council’s common position in April 2010 and the Parliament’s adoption of this position in second reading, the negotiations in trilogues in the late part of 2009 meant that an informal agreement had been reached already in December 2009 – considered a success for the Swedish Presidency (interview 6).91 The special facilitating role of the Presidency means that the level of pro-climate stakeholder involvement in the Council at this time is estimated as rather low (rather than very low), even though there was no visible involvement of the environment Council.

As much of the final negotiations took place behind closed doors in trilogues, among the Presidency, the rapporteur and shadow rapporteurs from the Parliament, and the Commission, it is difficult to assess accurately the interaction of pro-climate stakeholders throughout the policy process. One interviewee emphasised that the Parliamentary representatives are often at a disadvantage in trilogues, due to various reasons, including lacking negotiation skills, resources and expertise on the topic, among others (interview 5). On the EPBD recast dossier, the Parliament may have held the most ambitious “environmental” stance, but in the name of agreement and compromise, many of the Parliament’s green voices were left outside during the trilogue negotiations.

Overall, in the case of the involvement of internal pro-climate stakeholders in the policy process leading to the agreement on the recast of the EPBD, a low level of involvement can be ascribed (see Table 3.1). None of the traditional pro-climate stakeholders were in the lead in policy negotiations, although some of the pro-climate arguments may have in fact been taken over by other actors. Regular procedures were

91 The same interviewee commented that the desire to push the dossier through within the term of the Swedish Presidency may ultimately have resulted in a further watering-down of the piece of legislation for the sake of a speedy process (and to have the legislation agreed in time for COP15 at Copenhagen in December 2009).

192 CPI into EU energy performance of buildings policy in place to allow pro-climate stakeholders access to the policy process in the Commission and in the Parliament, but access was then limited in the Council and in the trilogue negotiations. Internal pro-climate stakeholders’ arguments were represented by some of the main policymakers, but the ENVI committee did not give an opinion on the dossier, and neither was it discussed in the environment Council. While internal pro-climate stakeholders may have been involved in early phases of the policy negotiations, the later negotiations in trilogues excluded all actors outside those directly negotiating (excluding internal pro-climate stakeholders).

External pro-climate stakeholders The involvement of external pro-climate stakeholders in the policy process leading to the 2010 EPBD recast is medium. Unlike in the lead up to the 2002 EPBD, environmental NGOs had by the late 2000s started to pay attention to the energy efficiency issue, and to invest in resources to influence the policymaking processes. Interviewees from the environmental NGO community in Brussels freely admitted that the lack of resources was a barrier that previously played against them in the development of policy on the energy performance of buildings. However, although energy efficiency was an issue that was taking more of their attention, they were still in a mode of learning, and still suffered from some resource constraints (interviews 5 & 6). Two interviewees also complained about difficulties in knowing the timetable of work of the Commission to be able to provide direct input at the right moment (interviews 5 & 11). As such, environmental NGOs were not much involved in the process leading to the Commission’s proposal for a recast of the EPBD in 2008.

In the run-up to the publication of the Commission’s proposal, several conferences and formal consultations with stakeholders took place (both online and in person; European Commission, 2008b, pp. 15-17). These consultation procedures allowed access for external pro-climate stakeholders in the drafting of the proposal stage in the policy process, so the limited involvement of environmental NGOs is due more to their own internal limitations than that they were blocked from the procedures. In its impact assessment accompanying the proposal for the recast of the EPBD, the Commission describes the consultations that took place in advance of the adoption of the proposal (European Commission, 2008b). Its online consultation took place for eight weeks from 25 April 2008, and received 246 responses, with organisations and citizens from 22 member states responding (ibid., p. 16). Fifty of these responses

193 CPI into EU energy performance of buildings policy came from NGOs, with another 33 from “associations”, while 83 responses were from industry. Institutions and member states provided 24 responses, while 44 responses came from citizens and the remaining five responses were filed under “other” (European Commission, 2008b, p. 4, Annex 1). Environmental NGOs were part of a catch-all “other category” for disclosing their field of activity that included 55 responses from organisations/individuals working in any of the following fields: environmental; energy; engineering; renewable and sustainable energy; energy efficiency auditing and certification; lighting; insulation; research; ventilation and/or heating systems; architecture; consultancy; products production; and trade (ibid., p. 5, annex 1). The Commission also highlighted other occasions where it interacted with member states and industry representatives to discuss options for revising the 2002 EPBD, such as concerted actions under the Intelligent Energy Europe programme in 2006-2007, and during a conference organised as part of the EU sustainable energy week in 2008 (ibid., p. 16). Thus, consultation procedures were in place and were used by environmental NGOs and pro-climate industrial stakeholders (and many other external stakeholders) leading up to the proposal of the Commission on the recast of the EPBD.

Many NGO representatives rather chose to invest their resources and effort into influencing the opinions of the Parliament, which interviewees considered as a more open institution (interviews 5, 6, 7 & 8). As one interviewee said, well-resourced environmental NGOs “can have an impact on the Commission… but that’s only before the proposal goes through the inter-service consultation… where it can then be changed completely”, whereas they “can access MEPs quite easily” and therefore can “have more of an impact on the Parliament” (interview 5). This ease of access in the Parliament led to what this particular interviewee described as the “best first reading ever” from an environmental NGO perspective, in terms of the improvements suggested through the amendments proposed by the Parliament.

Beyond the Parliament, environmental NGOs described great difficulty in influencing negotiations in the Council, and especially the trilogue negotiations that took place in the final months before the adoption of the Council position (interview 6). “These meetings [trilogues] are done behind closed doors”, complained one interviewee, “there’s a lot of lost transparency and it’s difficult to keep track of what’s going on” (ibid.). For this reason, environmental NGOs had little access to policymaking in the

194 CPI into EU energy performance of buildings policy

Council, and in the final stages of the negotiating process. Here, there are no procedures for ensuring access of external climate policy stakeholders to the policy process.

With regard to pro-climate stakeholders from industry (in this case, industries interested in pushing for ambitious energy efficiency measures), those industry associations involved in the first EPBD were also present in the negotiations leading to the EPBD recast. Pro-climate industry stakeholders were more organised at the EU level in the negotiations on the EPBD recast (Boasson & Wettestad, 2013, p. 142), and even linked with environmental NGOs in joint campaigns. The coalition for energy savings, for example, brings together 26 industrial associations and environmental NGOs to push for more ambitious policy measures on energy efficiency.92 It was hoped that the coalition could have a clear impact on the policy process, with industry and environmental voices presenting the same message (for more ambitious policy).

While environmental NGOs found themselves with better access to the Parliament, interviewees from industrial associations stated a clear preference for lobbying the Commission (interviews 10 & 11). The Commission, as one interviewee stated, is interested in “fact-based representation from industry”, while the Parliament is far more political, making it difficult for industry representatives to position themselves in lobbying the Parliament (interview 11). The Commission’s impact assessment for the EPBD recast proposal does also refer to studies in the OECD member countries, carried out by the IEA, and commissioned by organisations such as Eurima, EuroACE and FIEC (European Commission, 2008b; Geller, Harrington, Rosenfeld, Tanishima, & Unander, 2006). Working together in the coalition, then, may have had the benefit of ensuring the joint message from industry and environmental NGOs reached both the Commission and the Parliament.

92 As of March 2013, the members of the coalition for energy savings include: BPIE, ClientEarth, CAN-Europe, COGEN Europe, E3G, Energy Cities, PU Europe (insulation manufacturers association), European Copper Institute, EEB, European Federation of Buildings and Woodworkers, European Federation of Intelligent Energy Efficiency Services, European Partnership for Energy and the Environment, Friends of the Earth Europe, Glass for Europe, Lighting in Europe, Architects’ Council of Europe, EuroACE, European Alliance to Save Energy, ECF, European Committee of Domestic Equipment Manufacturers (CECED), the European Council for an Efficiency Economy, Eurima, the European Liaison Committee for Social Housing (Cecodhas), the Regulatory Assistance Project (RAP), the Royal Institution of Chartered Surveyors, WWF European Policy Office. See energycoalition.eu/members, accessed 19 March 2013.

195 CPI into EU energy performance of buildings policy

Interestingly, stakeholders clearly mention that at the time of the negotiations leading to the recast EPBD, there were “no major stakeholders opposing the EPBD”, but that it was rather reticence from member states that posed the main barrier to agreeing stronger legislation (interviews 5, 6 & 10). With the lack of access for pro-climate stakeholders to the policymaking process within the Council and the negotiations in the trilogues, stakeholders were at a disadvantage in terms of trying to influence the really reticent actors. Two industry organisations, Eurofuel and the European Partners for the Environment (EPE) jointly released a statement that lack of financial incentives would hamper the implementation of energy performance standards in buildings (ENDS Europe, 2010b). EuroACE regularly voiced its opinion on the need to improve the EPBD further (ENDS Europe, 2009c). The influence of such statements, especially in the later stages of the policymaking process seems limited, however. While industry and environmental stakeholders were positive about the improvements made in the recast of the EPBD, they nevertheless pointed out their dissatisfaction with the final output, and the insufficiency of the new measures to meet the (non-binding) 20 per cent energy efficiency target by 2020 (ENDS Europe, 2010a, 2010b).

Overall, the level of involvement of external pro-climate stakeholders in the policy process is medium (see Table 3.1). Procedures existed in the preparation of the proposal that allowed external stakeholder involvement (public and online consultations). Such procedures did not exist in later stages of the process, and stakeholders complained of lack of access to the Council and to the policymaking negotiations taking place in trilogues. Thus, procedures allowed the access of external pro-climate stakeholders, but the pro-climate lobby faced opposition from policymakers in the Council especially (although no major external stakeholder expressed opposition to ambitious policy measures for the EPBD). The Parliament’s first reading opinion was warmly welcomed by pro-climate stakeholders, but it was later substantially watered down during the trilogue negotiations. The internal Council negotiations and the trilogues turned out to be the venues for serious negotiations on this policy process, and external pro-climate stakeholders had no access to, or involvement in, these venues.

196 CPI into EU energy performance of buildings policy

Recognition of functional interrelations The recognition of the functional interrelations between climate policy objectives and EPB policy objectives in the policy process leading to the 2010 EPBD recast is medium. From the outset, and following on from the same framing in the 2002 EPBD, the Commission highlighted the environmental and climate benefits of improving the energy performance of buildings in the EU (along with benefits in terms of energy security and job creation; European Commission, 2008c, p. 12). The recast did not change the objectives of the EPBD, but rather aimed to clarify further concepts and tighten measures so that the EPBD could do more than simply bring energy efficiency onto political agendas, which is what the Commission considers to have been the main contribution of the 2002 EPBD (ibid., p. 3). Thus, the Commission follows the 2002 wording in the proposal for the EBPD recast on its potential to contribute to meeting the EU’s commitments under the Kyoto Protocol and to combat climate change generally, and also as part of the contribution to the non-binding energy savings target to 2020, but there is no explicit mention of long-term climate policy objectives.

In Parliament, there was a clear interest to push for ambitious policy on the energy performance of buildings. The first reading agreement proposed amendments to the Commission’s proposal that would have included wording about the EU’s long-term climate commitment to limit global temperature to 2°C in the opening recitals of the text. The Parliament also called for the 20 per cent energy efficiency target to 2020 to become binding; called for increasing the number of zero energy buildings in the EU; and proposed measures to increase financing for improving energy efficiency. The Council did not accept all these ambitious amendments, and the final EPBD outlined that new buildings would be “nearly zero energy” by 2020 and public buildings by 2018 – with the definition of “nearly zero” left to the discretion of the member states. Member states weakened Parliament’s proposals for financing and would not accept any binding efficiency targets (see also Boasson & Wettestad, 2013). The Parliament’s amendment to include the long-term commitment to limiting global temperature increase to two degrees Celsius remained throughout the negotiations (Recital 3, Directive 2010/31/EU). Yet this recognition of long-term functional interrelations was not backed up with the sort of ambitious policy measures that would then be required (i.e. the inclusion of Recital 3 did not translate into ambitious

197 CPI into EU energy performance of buildings policy energy efficiency measures to achieve 80 to 95 per cent reductions in GHG emissions by 2050, compared to 1990 levels).

In line with Table 3.1, the recognition of functional interrelations in the case of the 2010 EPBD recast is medium. While both short-term and long-term climate policy objectives were considered throughout the policy process, and are part of the stated motivations of the policy measure, the recognition of these functional interrelations did not lead to or motivate policy in an ambitious direction. While Council may be most at fault in this case, it may be that Parliament gave in to Council demands without considering the ramifications for long-term climate policy.

In summary, taking account of the three measurements outlined above, the level of CPI in the policy process is measured, overall and on aggregate, as low-medium (see Table 5.2).

5.3.2.2 CPI in the policy output Here, I attempt to determine the level of CPI in the policy output in the 2010 EPBD recast by establishing how much of the gap between BAU scenarios for energy consumption in buildings and very high or complete levels of CPI is closed by the policy output of this Directive (see section 5.2.2).

First, in its proposal, the Commission outlined that the potential for measures under the recast of the EPBD could amount to the avoidance of between 60 and 80 Mtoe energy consumption by 2020, compared to BAU (European Commission, 2008c, p. 4). With the disappointing results of the 2002 EPBD, the BAU scenarios had changed by this time. According to the 2006 action plan for energy efficiency, BAU scenarios (using DG Energy’s PRIMES modelling tool) foresaw 549 Mtoe energy consumption in buildings in 2020 (European Commission, 2006a, p. 6). The EPBD recast proposal could then be expected to reduce this level by between 60 and 80 Mtoe, to between 469 and 489 Mtoe. Very high or complete CPI means achieving a maximum of 200 Mtoe of energy consumption in buildings by 2050. With 2010 consumption standing at 483 Mtoe, this implies a steeper reduction in energy consumption than outlined above in Figure 5.4. Under complete CPI scenarios, therefore, energy consumption of buildings in 2020, starting from actual consumption rates in 2010, should be no more than 387-412 Mtoe. Thus, the proposal for the recast of the EPBD closes the gap of 137 Mtoe between BAU at 549 Mtoe and CPI of 412 Mtoe by between 44 and 58 per

198 CPI into EU energy performance of buildings policy cent. It closes the gap of 162 Mtoe between BAU at 549 Mtoe in 2020 and CPI of 387 Mtoe by between 37 and 49 per cent. This is represented in Figure 5.7. According to Table 3.2, this range of gap-closure from minimum 37 to maximum 58 per cent represents a low to medium level of CPI in the policy output of the EPBD recast.

Figure 5.7: CPI in the policy output of the 2010 EPBD recast, with measures expected to avoid between 60 and 80 Mtoe of energy consumption in buildings in 2020 (compared to BAU and very high/complete CPI range; measured in Mtoe).

Sources: European Commission, 2006a, 2008c; own calculations.

As discussed above in section 5.3.1.2 on the level of CPI in the 2002 EPBD policy output, other factors may temper the reality of achieving these objectives to 2020. These factors are similar to those that resulted in the very low impact of the 2002 EPBD, such as the lack of a binding target and poor implementation by member states. In the case of the 2010 EPBD recast, there was still no binding target for achieving energy efficiency measures, and fears were already being circulated at the agreement of the EPBD recast that it was unlikely the overarching 20 per cent energy savings by 2020 target would be achieved (Henningsen, 2011). The Commission later revised its estimates for the impact of the EPBD on energy efficiency generally in its energy roadmap to 2050 (European Commission, 2011g). The BAU scenarios in the roadmap include expected impact from the 2010 recast of the EPBD. Even with the EPBD recast in force, the energy consumption of buildings is still expected to rise by 2015 to nearly 500 Mtoe, before falling only incrementally to approximately 496 Mtoe in

199 CPI into EU energy performance of buildings policy

2020 (and only 472 Mtoe in 2050).93 Figure 5.4 above plots this BAU scenario to 2050 and Figure 5.8 below plots the updated BAU scenarios that include the impact of the EPBD recast to 2020 against very high/complete CPI trajectories.

Figure 5.8: Updated BAU scenario from 2011, which includes expected EPBD impact on the energy consumption in buildings to 2020, compared to the 2006 BAU trajectory and CPI ranges to 2020 (measured in Mtoe).

Source: European Commission, 2006a, 2011g; Heaps, et al., 2009; own calculations.

If we are to take Figure 5.8 into account, the level of CPI in the policy output of the 2010 EPBD recast changes. With 2006 BAU scenarios suggesting that energy consumption of buildings would reach 549 Mtoe, the new BAU scenario from 2011 that includes the impact of EPBD suggests the EPBD recast would avoid the consumption of 53 Mtoe in 2020. This closes the gap between complete CPI trajectories by between 33 and 38 per cent. Thus CPI in the policy output could be measured as low, according to Table 3.2.94

5.4 Explaining CPI in EU EPB policy Table 5.2, below, summarises the results of the analysis of CPI in the policy processes and outputs of the 2002 EPBD and its 2010 recast. The Table shows no evolution in

93 This is the average expected level across the five BAU scenarios presented by the European Commission in its 2050 energy roadmap (European Commission, 2011g).

94 Considering the poor implementation of the 2002 EPBD, a low level may prove a best case for the EPBD recast if member states follow a similar pattern of inadequate implementation.

200 CPI into EU energy performance of buildings policy the overall (aggregated) levels of CPI over the two time phases, although there is some small variation among the individual indicators of the level of CPI in the policy process in 2002 and in 2010. In the process leading to the 2010 EPBD recast, external pro-climate stakeholders had a higher level of involvement in the policy process than in the lead-up to the agreement on the 2002 Directive. This change was not significant enough to alter to overall level of CPI in the policy process, however.

CPI in the 2010 EPBD CPI in the 2002 EPBD recast CPI in policy output Low Low Internal pro- climate Low Low stakeholders: CPI in External pro- the climate Low Low to Medium Low to policy stakeholders: medium medium process Recognition of functional Medium Medium interrelations: Table 5.2: Levels of CPI in the policy process and output of the 2002 and 2010 EPBDs. In this section, I apply the four explanatory variables described in chapter three to the story of CPI in the 2002 EPBD and its 2010 recast to try to understand why such levels of CPI were found in each case. The four explanatory variables include functional interrelations; political commitment; the institutional and policy context (each of which can explain the level of CPI in the policy output and the policy process); and the process dimension (which can explain the level of CPI in the policy output only). The variables are discussed below with the view that their influence in the process could lead to influence in the output (as output follows from process). Thus I do not discuss separately the effect of each variable on the level of CPI in the policy process and the policy output, as these two are linked (Briassoulis, 2005a).

5.4.1 Functional interrelations First, I examine the nature of the functional interrelations between the EPBD’s objectives and long-term climate policy objectives. Where functional interrelations

201 CPI into EU energy performance of buildings policy exist, some demand for CPI is expected. In addition, functional interrelations may be more or less direct or indirect; and more or less synergistic or conflictual (see Table 3.4). In the EPB sector, functional interrelations are direct and synergistic. The EPBD objectives include reducing energy consumption in buildings. As less energy is consumed, less GHG emissions are emitted through the generation and transportation of energy, thus helping to combat climate change. Thus, the nature of the functional interrelations between the EPBDs and climate policy is considered direct – in other words, there are clear and close interrelations between the policy objectives – and they are considered synergistic – in that the policy objectives of both sectors are harmonious. Any scenario towards decarbonisation by 2050 highlights the importance of energy efficiency measures.

To achieve long-term climate policy objectives of reducing GHG emissions in the EU by between 80 and 95 per cent by 2050, energy consumption in the EU must decline. The EPBDs are part of this story, and should contribute to meeting this goal. In the case of the 2010 EPBD recast, synergies were especially emphasised for the benefit of achieving climate objectives. Not only were policy measures in the EPBD about reducing energy consumption in buildings, but there were also measures included to promote the use of renewable energy for any remaining energy consumption of nearly zero energy new buildings (Arts. 6, 9), meaning that GHG emissions in new buildings, at least, ought to be eliminated under EPBD provisions.

The direct and synergistic nature of the functional interrelations between policy to promote energy performance standards in buildings and long-term climate policy objectives is overarching for the 2002 EPBD and the 2010 EPBD recast. As outlined in Table 3.4, when the functional interrelations between the two policy areas are both direct and synergistic, most favourable conditions exist for the promotion of CPI. This reality could have helped shape demand for CPI in the policy process and can help advance levels of CPI. As is evident from Table 5.2, the nature of the functional interrelations was insufficient to advance CPI.

5.4.2 Political commitment When discussing the explanatory role of political commitment, I assess, first, the political commitment of the EU and its institutions to combating climate change in an

202 CPI into EU energy performance of buildings policy overarching sense, and, secondly, political commitment to promoting CPI in other policy sectors.

First, with regard to overarching political commitment to combating climate change in the EU, the story is the same across the case studies, with medium levels of political commitment in the early 2000s and high levels of political commitment in the late 2000s (see also chapters two, four and six). In the early development of energy efficiency policy at EU level, especially in the 1970s and 1980s, the major rationale was to decrease the dependence of the EU on imported supplies of energy (hence increasing energy security). There was no real political commitment to combating climate change at this time. However, with climate change rising on the international political agenda, in tandem with the agreement on the UN Framework Convention on Climate Change (UNFCCC) in 1992, and its Kyoto Protocol in 1997, the EU began tentatively to place climate change higher on its political agenda (Wurzel & Connelly, 2011a). Combating climate change began to feature more in the political discourse of the EU, and, from the 2000s, in the conclusions of the Council and the European Council (Oberthür & Dupont, 2011, p. 82). EU commitment to combating climate change in the late 1990s and early 2000s, when the EPBD was being developed, was rather more symbolic than concrete (see also chapter two). The EU repeatedly mentioned at this time its ambitions to be a world-leader in the fight against climate change, but had not developed internal policies to demonstrate concrete expression of these ambitions. In the early 2000s, and as demonstrated also in chapter four, EU political commitment to combating climate change at this time is thus considered medium in line with the indicators outlined in the threefold scale in Table 3.5.

Political commitment to combating climate change at the time of the negotiations on the 2010 EPBD recast was high. The EPBD recast was proposed in late 2008 (November) and negotiated throughout 2009. With the European Council having committed the EU to binding 2020 targets on GHG emissions reductions and renewable energy (see also chapters two and four) in 2007 (European Council, 2007), and to long-term commitments to 2050 in 2009 (European Council, 2009b), this period of time could be considered a high point in the overarching political commitment to combating climate change in the EU (Oberthür & Dupont, 2011). The negotiations on the EPBD recast also took place in the wake of the agreement on the

203 CPI into EU energy performance of buildings policy climate and energy package of policy measures (December 2008) and in the run up to the important Copenhagen climate negotiations in December 2009 (COP15). Political commitment to combating climate change generally in the EU was high at this time, although it was under some pressure due to the economic and financial crises that began in 2008 (and later also due to the perceived failure of the Copenhagen climate negotiations; Falkner, Stephan, & Vogler, 2010). As outlined in Table 3.5, a high level of political commitment means that statements of commitment are backed up with concrete policy measures – as is the case with the adoption of the climate and energy package at this time.

Secondly, I assess political commitment to CPI leading to the adoption of the 2002 EPBD and its 2010 recast. The Commission framed the 2002 EPBD as part of the EU’s response to fulfilling its commitments under the Kyoto Protocol to reduce GHG emissions in the EU (EU15) by 8 per cent by 2008-2012 compared to 1990 levels. Additionally, the Commission included a Recital in its proposal for the EPBD specifically referring to the legal objective of the EU to integrate environmental concerns into other sectoral policies. This remained throughout the negotiations and is the first Recital of the final Directive.95 The Commission demonstrates medium levels of political commitment to CPI in the EPB in 2002, both through statements on its motivations for policy development and through concrete inclusions of CPI in the policy proposal. The Commission is interested in advancing CPI through the EPBD, but seems to balance this against other objectives, such as energy security. It could also be argued that the level of political commitment of the Commission is tempered by a policy proposal on EPBD that is less than ambitious, including a lack of clear binding commitments.

The Parliament demonstrated medium levels of political commitment to CPI. It was generally in line with the Commission’s stance to achieve co-benefits in energy security and climate change, but raised concerns on the “third pillar” of the EU’s energy policy, namely competitiveness. The greatest concern in the Parliament was

95 Recital 1 of Directive 2002/91/EC states:

“Article 6 of the Treaty requires environmental protection requirements to be integrated into the definition and implementation of Community policies and actions”.

Article 6 of the Treaty establishing the European Community (TEC) is now Article 11 of the Treaty on the Functioning of the EU (TFEU; see Annex II).

204 CPI into EU energy performance of buildings policy the issue of financing energy efficiency measures, and this particular issue garnered much attention throughout the process. In the ensuing negotiations with Council, the timeframe for the implementation of the EPBD provisions was also at stake. Eventually a compromise was found that favoured the stance of the Council to delay implementation of the provisions of the Directive longer than originally proposed by the Commission. With regard to the political commitment to CPI in the Parliament, it seems that CPI received no special promotion, but rather that Parliament wished to promote integration of the three energy policy objectives in the EPBD – a difficult task in the ensuing negotiations with Council. Parliament had no objection to the Commission’s reference to Article 6 of the TEC on the integration of environmental protection measures. Political commitment of the Parliament to CPI is rather medium (see Table 3.5), but the commitment over the course of the policy process can be considered to wane, especially in the later negotiations with the Council.

The Council, however, demonstrates low levels of political commitment to CPI. It was far more reticent in terms of real ambition to integrate CPI into the EPBD and thus to help combat climate change through the use of the EPBD. Member states did not place combating climate change through energy efficiency high on their national political agendas, and argued forcefully for delayed implementation and extra flexibility. Commentators later mentioned that “the fact that implementation… [was left] to a great extent to member states” is a major reason for the overall failure of the 2002 EPBD in having any great impact in reducing GHG emissions in the EU (Henningsen, 2011, p. 133). Member states were concerned about the financial implications of the policy measure. They were also against any attempts to override the principle of subsidiarity, and opted for maximum flexibility for member states themselves to define the measures to be implemented (Boasson & Wettestad, 2013). Although the Council did not negotiate on the reference in the Commission’s proposal to Article 6, TEC, they weakened the Commission’s proposal and the Parliament’s amendments in many other areas that led to a policy instrument that would have little to no impact on GHG emission reductions (see also Figure 5.6). In the Council, political commitment to CPI through the EPBD can only be considered low (see Table 3.5).

In the case of political commitment to CPI in the EU institutions, we have a mixed picture in the 2002 EPBD. Taking the three co-deciding institutions together, an

205 CPI into EU energy performance of buildings policy overall and aggregated level of political commitment to CPI in the energy performance of buildings sector during this time phase is low to medium, but closer to medium. In the 2002 EPBD case, however, it could be that the Council’s low levels of political commitment hold more weight than the medium levels of political commitment in the two other co-deciding institutions.

With respect to political commitment to CPI in the 2010 EPBD recast, this is tempered by the lower level of political commitment to advancing measures on energy efficiency generally. Of the 2020 targets agreed by the European Council in 2007, only the target to save 20 per cent of energy by 2020 compared to BAU was left as an indicative or non-binding target. The European Council and the Council showed little ambition for strong energy efficiency measures, and thus there is little evidence of promoting CPI into the EPBD. Political commitment to promoting ambitious policy measures under the EPBD in the Council was low and unchanged from the 2002 EPBD. Member state concerns for flexibility, financial costs and considerations of national differences meant they had little appetite for pushing for ambitious EU- level policy for the EPBD recast. The one exception to this low level commitment may have been Sweden, which held the Presidency in the second half of 2009 in the lead up to the Copenhagen climate negotiations. However, in this case, it seems that Sweden was rather more interested in reaching agreement before the end of its Presidency term than in ensuring ambitious policy measures for combating climate change (interview 6; Boasson & Wettestad, 2013, p. 146).

In terms of political commitment to CPI, the Commission fares somewhat better than the Council, with medium levels. However, as the Commission’s proposal did not provide provisions to go far beyond the original 2002 EPBD, it was not considered ambitious enough by several stakeholders (interviews 5, 6, 7 & 10). It was in the Parliament’s first reading amendments that a more ambitious proposal came out, with environmental NGOs considering this one of the “best first readings ever” from a climate perspective (interview 5). The Commission was still in favour of ensuring the co-benefits of energy security and climate objectives were met in the EPBD, but with less commitment to proposing policy that would significantly contribute to the reduction of GHG emissions in the buildings sector within an adequate timeframe.

206 CPI into EU energy performance of buildings policy

In the Parliament, political commitment to CPI overall is also rather medium. Political commitment to achieving highly ambitious policy in EPBD in order to achieve the potential to reduce GHG emissions in the buildings sector was initially rather high in the first reading of the policy process. However, as the negotiations progressed, the Parliament compromised (in trilogue negotiations) on many of the more far-reaching proposals and agreed on a rather watered down Directive, with no timetables or binding targets, and few measures to improve the energy performance of the existing building stock. In some respects, the Parliament was hampered by the fact that all the implementation and standard setting of the policy measures in the EPBD were to be determined by the member states. It became clear in the second reading that the role of the Parliament was limited in face of the Council. Without the political commitment of the member states, even this “modified Directive…is unlikely to have much impact on energy consumption in buildings” (Henningsen, 2011, p. 134). Thus, Parliament’s political commitment to CPI in the 2010 EPBD recast can be measured as medium, in line with Table 3.5, where initially it stated its commitment to ensuring ambitious policy measures in the EPBD recast for combating climate change, yet was forced to give in on the ambitious policy measures (such as in terms of timetables, binding targets, renovations, and net zero energy standards) in negotiations with (an admittedly reticent) Council.

For the 2010 EPBD recast, the EU institutions’ (overall and aggregated) political commitment to advancing CPI in the EPBD policy was again low to medium. Although political commitment to CPI in both the Commission and Parliament scored a better medium level, the fact that the policy measures agreed need to be implemented without binding targets by mostly unenthusiastic member states suggest that the low level of political commitment to CPI demonstrated by the Council needs to be highlighted. In the case of EPB policy, the Council plays a crucial role in the level of ambition, it seems.

5.4.3 Institutional and policy context There are several internal and external institutional and policy contextual factors that may also help explain the level of CPI in EPB policy.

First, the institutional context at the time of the negotiations on the 2002 and 2010 EPBDs may shed some light on why such levels of CPI were found. Both the 2002

207 CPI into EU energy performance of buildings policy

EPBD and the 2010 EPBD recast were agreed under the co-decision procedure. The 2002 Directive was proposed and adopted under the environmental chapter of the Treaty, with Article 175(1) as the legal basis. The 2010 EPBD recast was also proposed under Article 175(1) TEC, but was finally adopted under the energy chapter of the TFEU, with Article 194(2) as the legal basis, as the TFEU was adopted in December 2009.96 The environmental legal basis may be seen as evidence of CPI, but before the TFEU, there was little competence for the Commission to make policy proposals in the energy field, except by taking an environmental or internal market legal basis (see also chapter two).

With the EPBDs being negotiated under the co-decision procedure, most decisions in the Council had to be made under the qualified majority voting rule (QMV).97 This meant that no single member state could veto a decision. As discussed also in chapters two and three, QMV may in some cases help open opportunities for policy change or allow for more ambitious policy measures. In this case, QMV does not seem to have had this effect. Throughout the negotiations on the EBPD, it was the member states that watered down the policy measure. Even member states who had implemented energy performance standards domestically (such as Denmark) did not seem to be pushing for ambitious policy at the EU level – there was little perceived incentive for member states to work together on this issue (with no “market” in buildings across borders, and questions of subsidiarity being raised). Leading to the 2010 Directive, almost all member states were either against strong measures for the recast, or else unenthusiastic or uninterested. This lack of interest may be linked more to the broader policy context of the buildings sector, rather than institutional issues. In this case, the procedural and decision-making context does not seem to play a role in advancing CPI – QMV decision-making rules did not advance policy change in favour of CPI, or did not present favourable conditions for advancing CPI. Thus, this institutional context can be considered as neutral for both the 2002 and 2010 EPBDs.

The long and slow development of EPBD policy in the EU could be an institutional contextual factor affecting the level of CPI. This long history includes the

96 See Annex II for the text of these Articles.

97 When Council votes in favour of an amendment that received a negative opinion from the Commission, then the Council must vote under unanimity rules.

208 CPI into EU energy performance of buildings policy

Commission trying to push for energy efficiency measures in the EU since the 1970s as a response to the oil crises, with limited success. While energy security motivations kept the attention of member states for some time, the energy efficiency issue eventually faded from member state priorities. The Commission continued to propose measures, however. Its 1984 communication on the rational use of energy in the buildings sector (European Commission, 1984), which made no reference to climate or environmental objectives, especially set the frame for the future policy proposals on the EPBD. The measures suggested by the Commission in 1984 are similar to those proposed finally for the EPBD (European Commission, 2001d).98 The failure of the 2002 EPBD to have any palpable impact on the energy consumption of buildings also pushed for the 2010 EPBD recast. However, even the lessons of the failed 2002 EPBD did not push the Commission to alter the framework of the policy measure. The 2010 recast proposal kept the general objectives, measures, and framing of the 2002 EPBD and kept the member states as the main players in the success (or otherwise) of the ensuing policy measure. With the EPBD framework traced back to the 1984 communication, this demonstrates a certain path dependency in the development of the EPBD, where past policy directions determined the sort of policy measures in the EPBD. This particular context could have proved unfavourable, if past policy weaknesses remained a part of the new policy developments. Such a context allowed little room for radical policy change (especially combined with the arguments for subsidiarity by member states), and suggests that there may have been few opportunities to push policy to advance CPI.

During the course of the 2000s, the EU underwent two rounds of enlargement – in 2004 and in 2007 – meaning the EU grew from 15 to 27 member states. The 2002 EPBD was negotiated and agreed among 15 members, while the 2010 EPBD recast involved 27 member states. Although the Eastern European newer member states were very much against strengthening the EPBD during the negotiations on the recast, they certainly were not alone among member states in their opposition. The Council was generally reticent to support ambitious policy measures, even when it was composed of just 15 members. However, it is clear that these new member states presented a particular challenge for reaching agreement when it came to proposals for

98 I.e. the promotion of energy auditing/certification; technical improvements; use of financial sources; information provisions (certification); regulations on minimum energy standards (European Commission, 1984, p. 4).

209 CPI into EU energy performance of buildings policy measures in the existing building stock. The final Directive is weak on measures for improving the energy performance of the building stock, with no concrete renovation plans outlined, and only states that buildings that happen to be undergoing “major” renovations should ensure minimum energy performance standards are met (Recitals 15, 16; Arts. 1, 7).99 As such, the enlargement of the EU presents a context that is rather more unfavourable for advancing of CPI throughout the policy process and in the output in the later 2010 EPBD recast.

Additionally, a peculiarity of the 2010 EPBD recast is that the negotiations on this file took place in the context of discussions on the EU’s energy policy more broadly (with the second strategic energy review having also been published in 2008; European Commission, 2008d), and in the wake of the agreement on the climate and energy package. While it is difficult to pinpoint interactions between the many energy and climate policy negotiations and discourses going on at that time, there may have been some influence of renewable energy policy on the 2010 EPBD recast. Perhaps ensuing from the renewable energy (RE) Directive discussion (see also chapter four), the 2010 EPBD recast includes measures that nearly zero energy new buildings should promote energy from RE sources. One direct link between the RE Directive and the 2010 EPBD recast includes the presence and active involvement of Green MEP Claude Turmes in both policy files (as rapporteur for the RE Directive and as shadow rapporteur for the EPBD recast; Boasson & Wettestad, 2013). This seems to suggest that this contextual factor is favourable towards advanced levels of CPI.

Second, the policy context throughout the first decade of the twenty-first century may explain part of the story. This includes the nature of the building sector itself, as well as international developments on climate change, and later, the financial and economic crises. When it comes to the building sector in the EU, it is characterised by high diversity and local level action (Boasson & Wettestad, 2013). The building stock in different member states is of different ages, types, and energy performance standards, with much of the industrial activity in the sector being small to medium enterprises, operating at a local or regional level (with very little interaction across borders in the EU; BPIE, 2011; European Commission, 2013b). This reality makes

99 The 2012 energy efficiency Directive (EED) does require member states to develop building renovation roadmaps, which may fill some of the policy gap on the building stock (Directive 2012/27/EU). The EED is not assessed in this chapter.

210 CPI into EU energy performance of buildings policy

EU-level policymaking problematic, and concerns about the EU being the appropriate level for policymaking on this issue have been expressed both within the Commission and from member states over time. This issue of subsidiarity resurfaces regularly, in fact, and is probably one of the main issues behind the lack of ambition on EPB policy in the EU. It is plausible that it is because of this reality, and because of the subsidiarity concerns, the Commission could not dare to propose radically ambitious policy (for example, with binding timetables and targets). This context is rather more unfavourable towards advancing ambitious policy measures and to promoting CPI in EU EPB policy.

A second policy context was rather more favourable towards policy development in this field – namely the ongoing international negotiations on climate change. The EU ratified the Kyoto Protocol in 2002, and during the negotiations on the EPBD, the contribution of energy efficiency measures was considered important for meeting the EU’s goals under the Kyoto Protocol. This is interesting as a response to the global climate negotiations, as energy efficiency itself is not a topic directly addressed in the international climate regime (see also Boasson & Wettestad, 2013, p. 130). Nevertheless, the international climate negotiations and the Kyoto Protocol helped provided additional motivation for the Commission to propose EPBD policy measures in the early 2000s. The EPBD recast was negotiated at a time of heightened activity in the field of climate change internationally, and of great EU ambition to demonstrate leadership on the issue. While the EU had expressed its leadership status on the climate issue over the years, it had barely begun to follow through with concrete targets and legislation internally to add credibility to these claims (Oberthür & Roche Kelly, 2008; Wurzel & Connelly, 2011). The EU considered its internal legislation as a way to demonstrate international leadership by showing its commitment to taking action on climate change itself. The Swedish Presidency managed to reach political agreement on the EPBD recast in November 2009 in time for the COP15 negotiations in Copenhagen. Paradoxically, the motivation to agree on the legislation (at least politically) quickly may have pushed through more compromises due to time pressure (interviews 5, 6 & 7). This time pressure may have produced unfavourable conditions for CPI in the 2010 EPBD recast, yet generally there is a motivating element associated with the international climate negotiations throughout the time period 2000

211 CPI into EU energy performance of buildings policy to 2010. Thus, this context may be considered to play a more favourable role in advancing CPI in EPB policy in the EU generally between 2000 and 2010.

The particular economic context during the negotiations on the 2010 EPBD recast is unfavourable to advancing ambitious policy and to advancing CPI. The 2010 EPBD recast proposal was published in 2008, just at the beginning of what turned out to be major financial and economic crises. In this context, negotiations in 2009 on financing measures for the energy performance of buildings were particularly challenging. More optimistic voices tried to highlight the job creating benefit of ambitious EPBD policy measures (interviews 5 & 10), but finally no specific financing targets were agreed. Article 10 leaves all discretion on the amount and nature of any financial incentives to the member states.

Taken together, however, there are more institutional and policy contextual issues that are unfavourable towards ensuring higher levels of CPI in the policy output and process of both the 2002 EPBD and the 2010 EPBD recast. It seems that the reality of the building sector in the EU, combined with concerns for subsidiarity and member state reticence were unfavourable in both policy phases. The 2010 recast faced the added problem of being negotiated during the economic and financial crises that began in the autumn of 2008. The more favourable context of the international climate negotiating agenda was insufficient to balance these several unfavourable institutional and policy contextual factors in the 2002 and 2010 EPBD negotiations.

5.4.4 Process dimension The policy output of a particular policy measure is preceded by the policymaking process involving negotiations on aspects of a policy proposal. Thus, for an analysis of CPI, the links between the policy process and output can provide clues as to the level of CPI found in the final policy output.

In the EPBD cases, the level of CPI in the policy process of both the 2002 and 2010 EPBDs was found to be low-medium, while the level of CPI in the policy output in both cases was found to be low. The levels of CPI found in the policy processes are overall and aggregated results of three indicators: the involvement of internal pro- climate stakeholders in the policy process; the involvement of external pro-climate stakeholders; and, the recognition by the policymakers of the (existence and nature of

212 CPI into EU energy performance of buildings policy the) functional interrelations between the EPB policy objectives and climate policy objectives to reduce GHG emissions by between 80 and 95 per cent by 2050 in the EU. While in both the 2002 and 2010 cases the level of CPI found in the process was low-medium, there was a change over the same time period in the level of involvement of external pro-climate stakeholders. In the negotiations leading to the 2010 EPBD recast access to the process was easier for external stakeholders and more stakeholders were involved and visible in the policy process than in the negotiations on the 2002 EPBD. However, this change was insufficient to improve the overall levels of CPI in the policy process, or, consequently, in the policy output (as these external stakeholders did not succeed in pushing for ambitious policy outputs). Additionally, policymakers initially recognised the functional interrelations between EPBD policy measures and climate policy objectives (especially for the short-term climate policy objectives under the Kyoto Protocol), but little attention was given at this time to long-term climate policy objectives throughout the process.

It is rather through breaking down the process dimension further that we can find more clues about the elements that advanced or blocked CPI. In both the 2002 and the 2010 EPBD, it seems that the Council played a crucial role in lowering the levels of CPI. The involvement of both internal and external pro-climate stakeholders in the policy processes in 2002 and 2010 was low. In 2010, stakeholders specifically mentioned the Council as the only clear blocker of ambitious policy for the 2010 EPBD recast (interviews 5 & 6). The arguments put forward of subsidiarity and the lack of either enthusiasm or interest of the Council seemed to outweigh other factors in the policy process and could plausibly thus be considered a main element of the process dimension that resulted in lower levels of CPI in the policy output also.

In summary, the process dimension can help explain the level of CPI in the policy output of both the 2002 and 2010 EPBDs. There was no overall change over the time period in the level of CPI found in the policy processes or outputs of these two Directives, but in breaking down the process dimension further, it seems that the Council played a significant role in both cases in lowering the ambition of the policy measures and in reducing the levels of CPI.

213 CPI into EU energy performance of buildings policy

5.4.5 Overview of explanatory framework Table 5.3 below provides an overview of each of the explanatory variables and how they can help explain the levels of CPI in the policy process and output of both the 2002 and 2010 EPBDs. There is little overall variation in the explanatory variables between 2002 and 2010, but Table 5.2 also showed little to no variation in the final levels of CPI in the processes and outputs of the 2002 and 2010 EPBDs. This lack of evolution is an interesting result for an energy efficiency policy measure, which, in theory, cannot be bad for the climate. How then is it possible that an energy efficiency measure shows low levels of CPI in the policy output in both 2002 and 2010 and low to medium levels of CPI in the policy processes in both time phases?

The only variation in the explanatory variables can be found in the level of political commitment overarching in the EU to combating climate change. This moved from medium to high between 2002 and 2010. However, this was clearly insufficient to push higher levels of CPI in EPBD policy. The nature of the functional interrelations are overarching both time phases examined here and are direct (i.e. they are clear and close together, with few steps between the interrelations of EPB and climate policy objectives) and synergistic (i.e. the objective of reducing energy consumption of buildings interrelates harmoniously with the objective of reducing GHG emissions) – thus more favourable towards higher levels of CPI. Yet, the nature of the functional interrelations seems also insufficient by itself to push up levels of CPI. It is, thus, important to look in more detail into the crucial blocking elements captured by the explanatory variables.

214 CPI into EU energy performance of buildings policy

Explanatory variables 2002 EPBD 2010 EPBD recast 1. Functional interrelations Direct & synergistic Direct & synergistic (process and output) 2. Political commitment (process and output) - To climate change Medium High

- To CPI Low-medium Low-medium

3. Institutional and policy context More unfavourable More unfavourable (process and output) 4. Process dimension Crucial role of the Crucial role of the (output only) Council Council Table 5.3: Summary of explanatory variables for CPI in the 2002 EPBD and the 2010 EPBD recast. In the process dimension, the role of the (generally reticent) Council is highlighted as one possible (crucial) explanatory element that could hold more weight over others. With regard to the level of political commitment to CPI in particular, it is again the Council that is the least committed to advancing CPI through EPB policy in both time phases (see above). The nature of the building sector, as a local or regional industry with little in terms of cross-border trade or interaction, also supports the notion that EPB policy is rather an issue for member state policymaking, backing up many Council opinions that subsidiarity rules would lead to questions about the need for EU level policy on buildings in the first place. Pro-climate policy stakeholders had little access to the Council and no major role in the negotiations in that institution. The recognition of the functional interrelations between EPB policy and climate policy objectives seemed limited to symbolic statements and there was no recognition of the interrelations with long-term climate policy objectives to reduce GHG emissions by 80 to 95 per cent by 2050. Each of these explanatory elements highlight the particular blocking element found in the institution of Council. The Council watered down policy proposals in both the 2002 EPBD and the 2010 recast, even when internal and external stakeholders were in favour of more ambitious policy. Without binding targets, member states retained their flexibility and discretion with regard to the policy measures.

215 CPI into EU energy performance of buildings policy

Much past literature on EPI in the EU highlights the pivotal role of political commitment in advancing policy integration (EEA, 2005; but see, e.g. Jordan & Lenschow, 2010). In this case, the levels of political commitment to advancing CPI seem insufficient, yet there is the added emphasis on the low levels of political commitment to CPI of the Council. Thus, when it comes to the pivotal role of political commitment, this case study seems to suggest that it is political commitment to CPI that is important, especially within the Council. But even at that, political commitment is an insufficient explanatory variable by itself. EPI literature also highlights the importance of the role of stakeholders in the policy process to advance policy integration (see, e.g. EEA, 2005). Yet the EPBD case shows that this is also an insufficient explanatory variable for understanding the levels of CPI. In the 2010 EPBD recast, in particular, stakeholders described no major lobby that was against strong policy measures in the EPBD to promote climate policy objectives. The Council was again pointed out as the main blocker of ambitious policy measures. Here, it seems that this case shows that policymakers cannot rely on the lobbying power of external stakeholders to push for more CPI in the process and output.

By breaking down the explanatory variables of political commitment and of the process dimension, a clearer understanding of the crucial role played by the Council in the lack of development of CPI between 2002 and 2010 becomes evident. The remaining explanatory variables can be seen as providing a background context that is either more or less favourable towards the advancement of CPI – such as the nature of the functional interrelations providing a favourable background context for CPI, while the institutional and policy contexts are more unfavourable to advancing CPI. These variables add extra layers to the overall story of CPI in the policy process and output of the 2002 and 2010 EPBDs.

5.5 Conclusions EU policy on the energy performance of buildings is particularly interesting for assessing the level of CPI in the EU’s energy sector. The EPBD and its recast were promoted as energy policy measures achieving a multitude of objectives, including improving energy security, combating climate change, and job creation. Since these measures have been in place, they have come to be regarded as climate policy

216 CPI into EU energy performance of buildings policy measures, and it is indeed difficult to imagine a policy measure on energy efficiency that does not help meet climate policy objectives.

In this chapter, I examined CPI in the policy output and process of the 2002 EPBD and of the 2010 EPBD recast. These policies included measures to improve the energy performance of buildings by improving the building envelope and the energy efficiency of installed equipment (such as hot water boilers) and by requiring the energy certification of buildings. Both the 2002 and 2010 Directives focused on measures to improve the energy performance of buildings in new buildings, and in buildings undergoing major renovations only. The 2002 Directive included a 1000m2 floor area threshold for building renovations that were obliged to meet minimum energy performance standards. In the 2010 Directive, this threshold was removed, and all buildings undergoing “major renovations” are expected to meet these minimum standards. Both Directives leave implementation and interpretation up to the discretion of the member states. Definitions of minimum energy performance standards, of nearly zero energy standards, and of what constitutes cost-optimal measures are not provided in the Directives, and member states have the flexibility to implement their own interpretations of these terms.100 Additionally, no binding energy efficiency targets were agreed in either the 2002 or 2010 Directives.

Although assumptions may exist that policy measures on energy efficiency can be considered climate policy measures, the analysis in this chapter clearly demonstrates the inadequate levels of CPI in the policy processes and outputs of the 2002 and 2010 EPBDs. With low-medium levels of CPI in the policy processes and low levels of CPI in the policy outputs of both Directives, these policies do not demonstrate sufficient levels of CPI to help achieve long-term climate policy objectives to 2050. The history of the development of the EPBD shows that combating climate change was a late motivation for policy development. Energy security concerns dominated energy efficiency policy objectives in the 1970s and 1980s, and this motivation remained part of the policy objectives after combating climate change was added. The 2002 EPBD was very much modelled on the policy measures outlined in the Commission’s 1984

100 Although guidelines are available from CEN, the European Committee for Standardisation, member states are not obliged to follow these guidelines.

217 CPI into EU energy performance of buildings policy communication on the rational use of energy in buildings, where neither environmental nor climate objectives are considered (European Commission, 1984).

The nature of the building sector and member state reticence for far-reaching policy measures at the EU level meant that the EU faced significant difficulties in adopting ambitious policy measures on improving the energy performance of buildings. The explanatory framework highlighted the importance, particularly, of the reticence of the Council as a barrier to advancing levels of CPI, but this explanatory element stood out against background contexts that could have been more or less favourable for the advancement of CPI. With regard to political commitment, in particular, the Council demonstrated the lowest levels of political commitment to CPI, bolstering the image emerging of the Council as the main barrier to CPI. While the nature of the functional interrelations provided a favourable context for CPI, the recognition of these functional interrelations did not focus on the long-term climate objectives to 2050.

These results again point to the importance but insufficiency of political commitment as an explanatory variable for CPI. Additional variables, when broken down, highlighted the role of the Council in particular, including arguments for subsidiarity, as a further dimension to the explanation. This case study may also point out the limited ability of external pro-climate stakeholders to influence a policy process when the Council shows itself to be unambitious – even when most external stakeholders lobby in favour of CPI and ambitious policy measures, the final policy output may be weakened by the Council.

218

CPI into EU external gas infrastructure policy

In this chapter, I explore climate policy integration in the policy process and output of the third case study, namely the EU’s policies in support of infrastructure for importing supplies of natural gas. This policy is one element of the EU’s external energy policy. To ensure a secure supply of natural gas, these policies aim to promote the construction of more natural gas importing infrastructure (hereafter “external gas infrastructure”), such as pipelines connecting the EU to natural gas producing countries.

The chapter opens with a discussion of natural gas in EU energy policy, including the uses and sources of the natural gas consumed in the EU, and the energy security and climate change issues around gas policy in the EU. This section also discusses the historical and expected development of domestic production and consumption, the sources and amounts of imported gas, and gas import infrastructure in 2010 and into the future. The second section moves directly to a discussion of the extent of CPI in the policy output and process of several EU-level policies related to external gas infrastructure. Here, I establish the benchmark “very high” or “complete” level of CPI against which actual policy is assessed. In section three, I first discuss the CPI in the external gas infrastructure policy of the EU’s trans-European networks for energy (TEN-E) policy field. Then, I turn to the EU’s Regulation on a European energy programme for recovery (EEPR), which was agreed in 2009. For both of these policy measures, I assess the extent of CPI in the policy process and output of their external gas infrastructure policy, with some anecdotal evidence from specific external gas infrastructure cases (such as the Nord Stream pipeline and the proposals for a Southern gas corridor). In section four, I explain the levels of CPI found in the EU’s external gas infrastructure policy using the explanatory framework outlined in chapter three, before concluding the chapter.

219 CPI into EU external gas infrastructure policy

6.1 Natural gas in the EU Natural gas is used in the EU mainly for heating, for some industrial processes, transport, and, increasingly, for electricity generation (European Commission, 2010d; IEA, 2011b). It is often promoted as the “cleanest” fossil fuel, with about half the greenhouse gas (GHG) emissions of coal (the fossil fuel that emits most GHGs).101 Its relative abundance globally,102 and the already proven and in-place technology and infrastructure to extract and distribute gas, make it an attractive option for the short- term transition period to decarbonisation. Its use as a “transition fuel” (or bridging fuel) has resulted in much discussion on the benefits and disadvantages of moving towards more gas-based electricity generation instead of coal (IEA, 2011b; Stephenson, et al., 2012).

Natural gas can be extracted and transported in different ways. “Conventional” natural gas is found in large deposits in gas fields and is extracted through usual drilling methods. It is generally transported by pipeline (unless it is converted to liquefied natural gas, LNG). LNG is natural gas that has been compressed and liquefied to be transported by ship. Shale gas, or “unconventional” gas, is extracted from shale rock formations using a method called “hydraulic fracturing” (or fracking) that involves pumping a mix of water and chemicals underground to fracture the rock in order to release and capture the gas deposits within these rock formations. For the purposes of this chapter, I focus more particularly on the development of pipeline infrastructure for transporting (usually) conventional natural gas, with reference also to the development of LNG infrastructure, as unconventional gas has not yet been developed in the EU (as of early 2013).

While conventional gas has a lower impact on the climate than coal, its use is nonetheless a major contributor to the climate problem. The different methods of extracting and transporting gas also result in different levels of GHG emissions. LNG, for example, has a greater impact on the climate, mainly due to the intensity of the liquefaction and regasification processes that take place, and the corresponding

101 Natural gas has often been promoted by the gas industry as a “clean” fossil fuel, suitable for the transition to a low-carbon economy (Harvey, 29 May 2012). This is despite the many studies questioning the “green” credentials of natural gas in all its forms (conventional, LNG and unconventional; Alvarez, Pacala, Winebrake, Chameides, & Hamburg, 2012; Howarth, Santoro, & Ingraffea, 2011; Lustgarten, 2011; Stephenson, Doukas, & Shaw, 2012; Wigley, 2011).

102 Global reserves of natural gas are estimated to last another 64 years at current consumption/production rates (BP, 2012), although with unconventional sources of natural gas becoming more important globally, these (low) estimates can be disputed.

220 CPI into EU external gas infrastructure policy shipping emissions from its transportation. Nevertheless, LNG has increasingly become an important element in the EU’s natural gas story over time, especially since the “shale gas revolution” in the US resulted in a supply glut of gas on the world markets, and corresponding drops in LNG prices (IEA, 2011b). Several EU member states already possess LNG import terminals, and there are more terminals planned, under construction or proposed (Gas LNG Europe, 2011; see section 6.1.3 below).

Shale gas is generally considered to have a higher rate of GHG emissions than conventional gas (with some commentators arguing that it can also be considered more climate damaging than coal, if emissions are measured on a shorter time horizon, such as 20 years instead of 100 years), but there is little agreement among scientists and industry about just how much shale gas impacts GHG emissions (Howarth, et al., 2011; Howarth, Santoro, & Ingraffea, 2012). In the EU, few countries have seriously considered shale gas as an option (the UK and Poland are among the member states investigating options for developing shale gas), and some have gone so far as to place a moratorium on permits for shale gas exploration until environmental and health impacts are more fully understood (such as France). Therefore, shale gas does not yet play a major role in the EU’s ambitions for energy independence and energy security (beyond the impact of the US developments on gas prices in the EU), and it certainly did not play a major role during the first decade of the twenty-first century – the time period under focus in this study.103

6.1.1 Natural gas and EU energy policy Although the European project can trace its origins to the energy policy of the European Coal and Steel Community in the 1950s, energy policy has long remained a competence of the member states. It is only since the entry into force of the Lisbon Treaty in 2009 that a new energy chapter entered the EU’s treaty system, making energy policy an official shared competence of the EU and member states (TFEU, Art. 4(2)).104 Title XXI on energy of the Lisbon Treaty (TFEU, Art. 194) provides a clear mandate for the EU in the area of energy infrastructure development. Article 194 of the TFEU reads as follows:

103 For general information on shale gas, see http://geology.com/energy/shale-gas/. Studies carried out for the European Commission on the energy market, environmental, health and climate impacts of shale gas can be found online here: http://ec.europa.eu/energy/studies/energy_en.htm, accessed 14 September 2012.

104 See Annex II for the text of Article 4(2) of the Treaty on the Functioning of the European Union (TFEU).

221 CPI into EU external gas infrastructure policy

“1. In the context of the establishment and functioning of the internal market and with regard for the need to preserve and improve the environment, Union policy on energy shall aim, in a spirit of solidarity between member states, to:

(a) ensure the functioning of the energy market;

(b) ensure security of energy supply in the Union;

(d) promote the interconnection of energy networks.”105

After long negotiations and several setbacks, the Lisbon Treaty entered into force in December 2009 (Devuyst, 2012). It remains to be seen how far this shared competence will allow decisions of energy infrastructure – traditionally linked to questions of structure of energy supply, which remains a member state competence106 – to be taken more at the level of the EU. Indeed, in the period 2000-2010, the entry into force of the Lisbon Treaty, coming at the end of the decade, cannot be seen to have had a major impact on developments. However, linking energy infrastructure decisions to overarching objectives of energy policy and to areas of environmental policy for which the EU has long held legislative power, helped this area of EU energy policy develop in the 2000s before the entry into force of the Lisbon Treaty (see also chapter two). Additionally, the Treaty chapter on trans-European networks (Title XVI, Articles 170-172, TFEU) provides further competence for the EU to support energy infrastructure development.107

The development of EU infrastructure policy can be linked to the objectives of ensuring sustainable energy development (linking renewable energy generation to the grid and updating the grid to handle variable energy supply, for example), competitiveness (by ensuring choice of energy supplier and type of energy) and security of energy supplies (linking to new sources of energy supply, increasing

105 See chapter two for more detail on the development of EU energy policy over time and see Annex II for the full text of Title XXI of the TFEU.

106 TFEU Article 194(2) states: “Such measures shall not affect a member state’s right to determine…the general structure of its energy supply…” (see Annex II).

107 See Annex II for the text of Title XVI, TFEU.

222 CPI into EU external gas infrastructure policy internal interconnections, developing new routes for energy exchange) – the three objectives of EU energy policy overall, as reiterated in the second strategic energy review (European Commission, 2008e, p. 2). In terms of sustainability objectives, it became clear that infrastructure developments would need to be developed hand-in- hand with RE generation. This infrastructure development was required for three main reasons. First, much of the RE is generated in areas that are remote from the load centres where the energy is consumed (e.g. coastlines, mountains, offshore, and often far from major consumption centres, such as cities), therefore requiring further infrastructure development to bring the energy to where it is consumed. Second, RE has long been considered as hazardous for the old electricity grid infrastructure in place in the EU. As RE generation is dependant on hours of sunshine and wind power (as the two predominant RE sources today), the grid needed to cope with variable supplies at times during the day and year that do not necessarily correspond with peaks of consumption. Third, the potential for major sources of different types of RE is dispersed geographically throughout the EU. Much potential for wind power generation, for example, exists in the North Sea, and in countries such as Denmark, Scotland and Ireland, among others. Solar power potential is greatest in the Southern countries of the EU. Bringing this energy to the places where it is required within the EU at large requires large cross border infrastructure projects to be put in place (European Commission, 2010d; see also chapter 4).

While sustainable energy ambitions have promoted EU-level electricity infrastructure policy, it is more the energy security objectives of EU policy that have promoted external natural gas infrastructure. The European Commission’s green paper in 2000 entitled a “European strategy for the security of energy supply” highlighted the need for a “stronger mechanism…to build up strategic stocks and to foresee new import routes” of natural gas supplies (European Commission, 2000d, p. 4). Policy measures on stocks and emergency responses to disruptions in supply were implemented with Directive 2004/67/EC and the follow-up Regulation (EU) No 2994/2010 on measures to safeguard security of natural gas supply. As they focus on short-term security measures, and very little attention in these pieces of legislation is given to longer-term security responses such as infrastructure development, they do not form the focus of analysis here. The two policy measures that directly provide support for external gas infrastructure include the trans-European networks for energy (TEN-E) guidelines

223 CPI into EU external gas infrastructure policy

(Decision 2003/1229 repealed by Decision 2006/1364, see section 6.3.2), and the later policy on the European energy programme for recovery (Regulation (EC) No 663/2009, amended by Regulation (EU) No 1233/2010, see section 6.3.3).

6.1.2 EU natural gas production, consumption and imports Domestic production of natural gas has been steadily declining in the EU. British Petroleum’s (BP) figures show a drop in production in the EU from approximately 232 billion cubic metres (bcm) in 2000 to 175bcm in 2010 (BP, 2011, p. 22), of which just 3.9bcm was exported to Switzerland and Turkey, meaning the great majority of EU-produced natural gas is consumed in the EU (BP, 2011, p. 29). Proven natural gas reserves in the EU stood at 2.3 trillion cubic metres (tcm) at the end of 2010, down from 3.8tcm at the end of 1991, resulting in an estimated 12 years of production remaining in these stocks (BP, 2012, p. 20). The main EU gas producing member states include: Denmark, Germany, Italy, the Netherlands, Poland, Romania and the United Kingdom.

Conversely, EU consumption of natural gas has remained at a relatively steady high level from 2000 to 2010, although it has faced some fluctuation. Gas consumption in the EU stood at 440bcm in 2000,108 and rose to 493bcm in 2010, but the peak of consumption was in 2005 at 494bcm (BP, 2011, p. 22; see Figure 6.1 and section 6.1.3 below). 109 Although Eurostat figures (measured in million tonnes of oil equivalent, Mtoe) show some slight variations to those reported by BP, the general trend is similar. The highest levels of gas consumption reported by Eurostat for the EU were also reported in 2005 and 2010.

108 This figure includes today’s EU27 member states. In 2000, the gas consumption of the then EU15 was around 382bcm (BP, 2002, p. 25).

109 In 2011, consumption dropped to 448bcm. This represented a 9.9 per cent drop in EU gas consumption in 2011 compared to 2010 – the largest drop in EU gas consumption on record (BP, 2012).

224 CPI into EU external gas infrastructure policy

Figure 6.1: Evolution of natural gas consumption and production in the EU from 2000 to 2010 (measured in billion cubic metres, bcm).

Source: BP, 2011.

Total natural gas imports to the EU in 2010 amounted to 421bcm110 (BP, 2011, p. 29), with approximately 80bcm being imported through the LNG network, and the remaining 341bcm through pipelines. This amount, combined with the amount of natural gas produced and consumed in the EU (c. 171bcm), exceeds the total consumption in 2010 by about 99bcm – gas that can be stored in facilities around the EU for periods of low supply/higher consumption.111 The drop in the level of domestic production, and the relatively constant rate of natural gas consumption has led to a situation of increasing EU dependence on natural gas imports. Eurostat has measured the dependence of the EU on foreign imports of natural gas over time, as a percentage. These figures reveal the increasing energy dependence of the EU. In 1998, the EU was nearly 46 per cent dependent on imports of natural gas to meet its energy demand. This rate grew steadily over the following years, reaching 58 per cent in 2005, and 62 per cent in 2010. In addition, several member states are 100 per cent reliant on imports of natural gas from various sources, including Sweden, Finland, Portugal, Luxembourg and Estonia. Many other member states are over 90 per cent

110 390.3bcm was imported in 2011 (BP, 2012).

111 Most EU member states are also locked into long-term contracts for receiving supplies of gas – another reason why supplies may exceed total consumption.

225 CPI into EU external gas infrastructure policy reliant on natural gas imports, including Slovakia, Slovenia, Lithuania, France, Spain, Greece, Ireland, Bulgaria and Belgium.112

EU member states import their natural gas from various sources and suppliers, both through gas pipelines and by LNG. In 2010, the sources of LNG into the EU included Qatar, Nigeria, Algeria, Trinidad & Tobago, Egypt, Peru, Equatorial Guinea, Yemen, the US, Libya, Oman and Norway. Of these suppliers, the most significant is Qatar, which supplied approximately 34bcm of a total of 80bcm of natural gas reaching LNG terminals in Belgium, France, Italy, Portugal, Spain and the UK in 2010, with Nigeria supplying just over 18bcm and Algeria supplying 8.5bcm of LNG (BP, 2011, p. 29). In comparison, LNG imports in 2001 amounted to 28.7bcm and were dominated by Algeria, providing 20.7bcm of gas to EU member states (BP, 2002, p. 28). In 2005, Algeria still dominated, providing 19bcm of the total 42.7bcm LNG supplied to the EU in that year, but with Qatar beginning to increase its LNG exports to the EU to nearly 5bcm (BP, 2006, p. 30), to the 2010 levels of nearly 34bcm (BP, 2011, p. 29).

Figure 6.2: Share of LNG supplies from different source countries into the EU in 2010.

Source: BP, 2011. Gas movements into the EU by pipeline in 2010 came from Norway, Turkey, Azerbaijan, Kazakhstan, the Russian Federation, Turkmenistan, Uzbekistan, Algeria

112 See Eurostat database, available at: http://epp.eurostat.ec.europa.eu/portal/page/portal/eurostat/home.

226 CPI into EU external gas infrastructure policy and Libya. In 2010, total pipeline imports to the EU were estimated at around 341bcm (BP, 2011, p. 29). The Russian Federation is by far the most important supplier of piped gas to the EU. In 2010, just over 110bcm of Russian gas was piped to Austria, Bulgaria, the Czech Republic, Estonia, Finland, France, Germany, Greece, Hungary, Italy, Latvia, Lithuania, the Netherlands, Poland, Romania and Slovakia (BP, 2011, p. 29). In 2001, Russian piped gas reaching EU member states amounted to nearly 113bcm (including piped gas to countries that were not yet EU member states at that time; BP, 2002, p. 28). In 2005, Russian piped gas imports to the EU member states amounted to about 128bcm (BP, 2006, p. 30).

6.1.3 EU natural gas import infrastructure According to the European Commission’s Energy Market Observatory data, EU import pipeline capacity totalled about 440.6bcm in 2010/2011. This includes direct pipelines from exporting countries to the EU, and transit through the pipeline networks of neighbouring countries such as Ukraine and Turkey. The Greenstream pipeline from Libya to Italy has the capacity to move 9.3bcm of gas per year (European Commission, 2010a, p. 11). In 2010, Algeria transported gas through four international pipelines to both Italy and Spain, with a combined capacity of 76bcm per year (European Commission, 2011c, p. 12). In 2011, Russia transported gas by pipeline directly to Estonia, Finland, Lithuania, Poland, Germany, Slovakia, Czech Republic, and onwards to other EU member states with a combined pipeline capacity of 136.1bcm (including the first phase of the Nord Stream pipeline connecting Russia to Germany, which became operational in November 2011, but not including pipeline capacity reaching Belarus, Ukraine and Turkey – through which gas is transited further to the EU; European Commission, 2010b, pp. 11-13).113 Finally, Norway transports natural gas to the United Kingdom, Denmark, Belgium, France and the Netherlands through several pipelines with a combined capacity of 144bcm (European Commission, 2011d, p. 12). Such pipeline capacity totals 365.4bcm directly connecting supplier countries to the EU in 2010/2011. Adding pipeline capacity from the (transit) pipelines in Ukraine and Turkey increases this total of pipeline import capacity to 440.6bcm in 2010/2011 (European Commission, 2010b, 2010c, 2012a).

113 See also: http://www.nord-stream.com/pipeline/, accessed: 15 December 2012.

227 CPI into EU external gas infrastructure policy

Several pipeline projects are planned or under construction from Norway (10bcm planned), Russia (63bcm planned through South Stream, 27.5bcm Nord Stream phase two pipeline constructed), and planned capacity through the Southern Corridor projects of either the Trans-Adriatic Pipeline (TAP) and/or Nabucco West (minimum 10bcm capacity) would increase the infrastructure pipeline capacity from 440.6bcm in 2010/2011 to about 551.1bcm per year by 2020, when all the planned projects are expected to be operational (European Commission, 2010b, 2011d, 2012a). As a point of comparison, in 2003 the European Commission mentioned that the total capacity of the then EU15’s natural gas importing infrastructure (including LNG) was about 330bcm per year (European Commission, 2003a, p. 7). BP reports that consumption levels in the EU15 in 2000 were about 376bcm (BP, 2002, p. 25), which corresponds to the higher levels of domestic production in the EU at the time.

LNG trade has been expanding globally, with a 10.1 per cent increase between 2010 and 2011 (BP, 2012, p. 4). The EU’s capacity to receive imports of LNG has also been expanding to reach a total capacity of 186.5bcm per year in 2011 (IEA, 2012, p. II.59). LNG terminals exist in Belgium, France, Greece, Italy, the Netherlands, Portugal, Spain and the United Kingdom. But this LNG capacity is planned for expansion with several current LNG terminals increasing their capacity, and several terminals under construction in Spain, Italy, France and Poland. Total capacity to import LNG should therefore reach about 274.4bcm by 2020, without taking into account project proposals (Gas LNG Europe, 2011).

Total gas infrastructure import capacity, therefore, combining pipelines and LNG terminals, amounts to about 626.5bcm for 2010/2011. Considering that the gas infrastructure in the EU is relatively “young” and such infrastructure is generally expected to have a lifetime of about fifty years (European Commission, 2010d, p. 5; European Parliament, 2009b) we can expect this capacity to remain in place to 2020, and much of it also to 2050. Adding the current capacity to the planned and under construction pipeline and LNG capacity that is due to become operational by about 2020, sees a total potential gas import infrastructure capacity in the EU of around 825.4bcm in 2020 (see table 6.1).

228 CPI into EU external gas infrastructure policy

Capacity in 2010/2011 Capacity by c. 2020

Pipeline capacity: 440bcm 551bcm

LNG terminals: 186.5bcm 274.4bcm

Total import capacity: 626.5bcm 825.4bcm

Table 6.1: EU gas import capacity 2010-2020, taking account of projects under construction, but excluding proposed projects such as the Southern gas corridor (measured in bcm).

Source: Adapted from Energy Observatory Market data; Gas LNG Europe, 2011; IEA, 2012.

6.2 Measuring CPI in EU external gas infrastructure policy In this section, I outline the benchmarks that I use to assess the level of CPI in the EU’s external gas infrastructure policies.

6.2.1 Benchmark for measuring CPI in the policy process As outlined in chapter three, and applied in chapters four and five, measuring the level of CPI in the policy process involves examining a number of indicators. This includes the involvement of internal pro-climate stakeholders in the policymaking process (such as DG Environment, the ENVI committee in the Parliament, and the Environment formation of the Council); the involvement of external climate policy stakeholders (e.g. environmental NGOs and pro-climate industry actors, such as renewable energy companies) in the discussions and consultations; and the recognition (and articulation) by decision-makers of the functional interrelations between the policy being developed and (long-term) climate policy objectives.

As with the earlier cases described in chapters four and five, very high levels of CPI in the policy process of external gas infrastructure policy would see very high involvement of internal and external pro-climate stakeholders and clear recognition of the functional interrelations (see Table 3.1). In this particular case, the recognition of the functional interrelations and very high involvement of pro-climate stakeholders could conceivably lead to the abandonment of any policy in support of external gas infrastructure. This is due to such policies supporting continued development of fossil-fuel infrastructure (with lifetimes of fifty years or more), which carries the risk of locking the EU energy infrastructure into a carbon-based system far longer than required under 2050 scenarios (see more below on the policy output).

229 CPI into EU external gas infrastructure policy

6.2.2 Benchmark for measuring CPI in the policy output Establishing the level of CPI in the policy output of EU policy supporting external gas infrastructure projects involves establishing what a very high or complete level of CPI would look like, and the distance between the status quo of the output and this very high level. What sort of policy output of EU external gas infrastructure policy would be in harmony with the 2050 goal of reducing GHG emissions by between 80 and 95 per cent compared to 1990 levels? Establishing this complete level of CPI in the policy output will then allow for an assessment of how far from that ideal level the actual policy output stands.

In this policy sector, a measurement of the extent of CPI (and thus an assessment of the necessity of supporting gas importing infrastructure) must take into account several interrelated elements for gas infrastructure policy development:

1. The expected natural gas consumption in a decarbonised EU in 2050; 2. The role to be played by carbon capture and storage technology (CCS) in the energy sector in 2050; 3. The amount of gas import infrastructure expected to be in place in 2050.

First, I will examine the link between CPI and expected levels of natural gas consumption in the EU to 2050 in a decarbonisation scenario. Decarbonisation scenarios to 2050 vary as to the amount of natural gas that will be required on the road to 2050 as a “back-up” fuel for intermittency of renewably generated electricity. Some scenarios claim that as little as 52bcm of natural gas will be required in 2050, and only for industrial processes (Heaps, Erickson, Kartha, & Kemp-Benedict, 2009, p. 19), while electricity and heating will be completely decarbonised by 2050. EREC also considers 100 per cent RE for 2050 a feasible reality for the EU (EREC, 2010), implying no need for natural gas in energy consumption (although some gas may still be required for specific industrial processes). Eurogas, unsurprisingly, promotes natural gas as a low-carbon energy source that could still provide as much as 462bcm in 2050, with CCS (Eurogas, 2011). The Commission’s energy roadmap to 2050 outlines a range of estimates for gas consumption as a total of the gross inland consumption in its 2050 decarbonisation scenarios from 233bcm to 320bcm. In these

230 CPI into EU external gas infrastructure policy scenarios, a minimum of 202bcm is used for gas fired power generation, 114 representing in each scenario the majority share of the use for natural gas in 2050 (the percentage use of natural gas gross inland consumption for gas fired power generation ranges from 73 to 94 per cent in the Commission’s 2050 decarbonisation scenarios; European Commission, 2011g, pp. 68-77).

It is reasonable, however, to expect natural gas consumption to rise slightly in the short-term (to, e.g., 2015 before beginning to decline) as more coal plants are closed on the road to decarbonisation. Additionally, since EU domestic gas production is decreasing, and reserves in 2011 were expected to last another 12 years approximately (BP, 2012), most natural gas consumption in 2050 will come from imports (excluding the potential for shale gas production in the EU, which is a discussion that goes beyond the scope of this chapter).115

Second, we must consider the role of CCS. None of the scenarios that foresee an amount of gas in the energy mix to 2050 can be compatible with long-term climate policy objectives without CCS, as full decarbonisation of the energy sector is required to achieve 80 to 95 per cent GHG emission reductions by 2050. The assumptions about the future of natural gas use in the energy mix to 2050 depend on the deployment of this technology for fossil fuel power generation. Shell’s 2050 scenarios outline that 90 per cent of all coal- and gas-fired power stations in the OECD would need to be equipped with CCS technologies in 2050 (Shell International BV, 2008, p. 32). The Commission’s 2050 energy roadmap makes the assumption that CCS will play a role in between 19 and 32 per cent of total power generation by 2050, depending on other assumptions such as shares of RE generation (European Commission, 2011a, p. 8), yet it states earlier in the same document that it remains “impossible to anticipate…whether and when carbon capture and storage (CCS) will become commercial” (p. 3). A 2012 appraisal of the state of the art of CCS technology outlines that it is not developing fast enough as a technology to play a significant role by 2050. Its role in 2050, under the current policy framework and current levels of development, would be “marginal at best” (Reichardt, Pfluger,

114 Figures converted from Mtoe, using BP (2012, p. 44) statistical conversion rates (1Mtoe = 1.11bcm).

115 Although, in 2013, there are discussions on whether or how to promote shale gas production in the EU, during the time scale under focus here (2000-2010), shale gas in the EU remained unexploited.

231 CPI into EU external gas infrastructure policy

Schleich, & Marth, 2012, p. 1). Informal assessments with interviewees suggest that CCS is becoming less and less feasible as time goes on, and one interviewee suggested that companies “are no longer interested in CCS” (interviews 17 & 19; IES Autumn lecture, 3 October 2012116). For the purposes of establishing the ideal level of CPI in the policy output, therefore, we cannot assume there will be a great roll-out of CCS technology, and it is the levels of natural gas consumption that must drop dramatically in order to achieve 2050 climate objectives.

Third, the amount of natural gas import infrastructure expected to be in place in 2050 should be considered when it comes to assessing the level of CPI in external gas infrastructure policy. As outlined in section 6.1.3 above, the capacity for natural gas imports in 2020 is likely to amount to 825.4bcm, not taking into account proposed projects, but only those projects operational or under construction. This adds nearly 200bcm of natural gas import capacity between 2010 and 2020 (see Table 6.1). As such infrastructure is generally expected to have a lifetime of about fifty years (European Commission, 2010d; see also nord-stream.com), this 200bcm of capacity will certainly still be in place in 2050, and even beyond. Considering that most EU gas import infrastructure is relatively young, much of the infrastructure available pre- 2010 is likely also to still be in place in 2050 (European Parliament, 2009b). An estimate of available gas infrastructure in the EU to 2050 can thus be between 600 and 800bcm, considering the infrastructure in place in 2010/2011, and the infrastructure under construction that should be ready by 2020. It should be noted that this estimate does not take account of the proposed projects (such as the Southern gas corridor, potential Nord Stream expansion, proposed LNG expansion that have not yet entered planning stages) that could potentially be built over the next decade and remain operational beyond 2050 (see Table 6.2). With investments in such infrastructure usually involving large sums (e.g. Nord Stream is estimated to have cost nearly EUR 8 billion to construct), there is a high economic pressure to continue operations at high capacity for economic return, representing a high risk of “lock-in” to this natural gas/carbon infrastructure.

Taking these three considerations (expectations of gas consumption under decarbonisation scenarios; the role of CCS; available import infrastructure in 2050), a

116 See: www.ies.be/ALS2012.

232 CPI into EU external gas infrastructure policy very high or complete level of CPI in EU external gas infrastructure policy output would involve decisions to limit expansion of infrastructure in line with efforts to reduce the consumption of natural gas (as a fossil fuel) to a minimum. Such a minimum level can be considered as including the consumption of sectors outside of power generation, where substitution of substances may be difficult. Thus, taking account of expectations for natural gas consumption in 2050 in gross inland consumption, excluding gas for power generation, and considering gas use in other sectors for, e.g., industrial processes, would lead us to a minimum figure of 18bcm and a maximum figure of 52bcm in natural gas consumption in 2050, assuming no gas fired power generation in 2050 – a rather radical scenario (European Commission, 2011g; Heaps, et al., 2009). This figure excludes Eurogas’ scenario of a major role for natural gas in power generation. Yet, current knowledge of the feasibility of a fully renewable energy generation system points to a danger of intermittency until storage technology is better developed (Giordano, Gangale, Fulli, & Sanchez Jimenez, 2011; von Hirschhausen, 2010). To take a more conservative stance on the role of gas in 2050 would thus include an amount of gas fired power generation. For the purposes of the analysis here, I therefore suggest up to a maximum of 150bcm of natural gas may be required by 2050, based on a survey of scenarios to 2050, and provided the limited number of gas fired power generation plants are equipped with proven CCS technology. Figure 6.3 compares such a CPI scenario with that of the business-as- usual scenarios (BAU, from 2005) and decarbonisation scenarios outlined by the European Commission’s energy roadmap. This figure clearly shows the distance between the ideal of CPI and the Commission’s decarbonisation scenarios.

233 CPI into EU external gas infrastructure policy

Figure 6.3: Range of very high or complete CPI gas consumption trajectories in the EU27 to 2050, compared to the BAU and decarbonisation scenarios of the Commission’s Energy Roadmap (measured in bcm).

Source: European Commission, 2011g; own calculations; Mtoe converted to bcm using BP conversion rate, with 1 Mtoe = 1.11bcm.

However, clearly the argument that natural gas may play a role in the short-term is valid as an alternative to coal, which is a more potent GHG emitter than gas. Figure 6.4 shows how continued high levels in natural gas consumption to 2015 is feasible (as we move away from coal plant power generation, and potentially moving away from nuclear generation also, considering the post-Fukushima context117). Such a reality will not, however, take away from the ideal level of CPI in 2050, which remains at between 18 and 150bcm. The effort in moving away from natural gas from 2015/2020 onwards becomes more pressing, with steeper consumption reduction ambitions required.

117 As of 2013, this shift away from coal and nuclear in the EU has yet to begin, with NGOs reporting an increase in coal-plants instead of a decrease (see, e.g., Greenpeace, 2013).

234 CPI into EU external gas infrastructure policy

Figure 6.4: Very high or complete range of CPI to 2050, with continued natural gas consumption to 2015 before consumption begins to reduce by 2020 in line with decarbonisation scenarios (measured in bcm).

Source: European Commission, 2011g; own calculations; Mtoe converted to bcm using BP conversion rate, with 1 Mtoe = 1.11bcm.

Table 6.2 outlines how such a level of CPI in natural gas consumption compares to the external gas infrastructure required. As shown above in Table 6.1, total import capacity in 2010/2011 stood at 626.5bcm, and capacity for 2020, taking account only of projects under construction, should reach 825.4bcm. Taking the conservative stance that only the infrastructure built between 2010 and 2020 will still be in place in 2050 (although more capacity is certainly likely) shows that infrastructure will already be more than sufficient to cover the higher ranges of natural gas consumption under very high or complete CPI levels in 2050 (see Table 6.2).

Thus, for EU external gas pipeline policy to be in line with 2050 climate objectives, it must be an EU-policy output that discourages investment in further (unnecessary) infrastructure. Policies that aim to support actively further gas importing infrastructure (whether politically, financially, or both) conflict with climate objectives, and thus CPI can only be non-existent in such a policy output. As discussed in section 6.3 below on specific EU gas importing infrastructure policies, policy output is measured on a scale from none/very low, low, medium, high, and very high/full CPI (Table 3.2).

235 CPI into EU external gas infrastructure policy

2010/2011 2020 2050 Total import capacity 626.5bcm* 825.4bcm 600-800bcm Very high CPI scenario: gas 493bcm** 336-380bcm 18-150bcm consumption Very high CPI scenario: with 493bcm** 427-447bcm 18-150bcm continued high gas consumption to 2015 *Actual imports in 2010 = c. 421bcm **Actual consumption in 2010 = 493bcm Table 6.2: Comparison of gas import infrastructure capacity to natural gas consumption expectations under very high CPI scenarios.

Source: BP, 2011, 2012; European Commission, 2011g.

6.3 EU policy on external gas infrastructure EU policy on the promotion of external natural gas infrastructure is most closely linked to objectives and discussions on improving the security of supply (see, for example, Bahgat, 2006; Baran, 2007; Brown & Huntington, 2008; Skea, Chaudry & Wang, 2012). Energy security has been one of the major motivating factors for the development of EU energy policy since the very beginnings of the European integration project. It was rather through the linking of energy policy measures, however, to policies on the environment and the internal market that more EU energy policy measures began to be adopted in the 1990s and 2000s (see chapter two). In its 1995 green paper on energy policy, the Commission outlined the “trinity” of EU energy policy goals that have remained central to EU energy policy since: security of supply, competitiveness, and sustainability (European Commission, 1995, p. 5). While policies to promote external gas infrastructure connecting third countries to the EU thus ought to consider the implications of each of these three policy objectives, the policy measures are situated most directly in the discourse on the security of supply.

In this section, I first discuss the security of gas supply context in the EU policy development over the course of 2000 to 2010. Next, I focus on the main policy measures decided at the EU-level on the promotion of external gas infrastructure in the same time period – namely the trans-European networks for energy (TEN-E,

236 CPI into EU external gas infrastructure policy

Decisions from 2003 and 2006), and the European energy programme for recovery Regulations (EEPR Regulation, adopted in 2009, amended in 2010). For each of these policy measures, I outline the development of policy on external gas infrastructure throughout the policy process, and discuss the level of CPI in both the process and output.

6.3.1 Security of supply context in the EU As noted above in section 6.1.2, EU natural gas production has been steadily declining, while consumption has remained constant or has risen slightly. This situation has led to an overall increase in the dependence of the EU on imports of natural gas, with Russia being the main supplier to the EU. The increasing dependence on external suppliers was not lost on policymakers in the EU between 2000 and 2010. The Commission put forward a proposal for a directive in 2002 on measures to safeguard the security of natural gas supply (European Commission, 2002c). Fears of weaknesses in energy supply security seemed to be realised in 2006 and 2009 when supplies of gas from Russia were cut for periods of time due to disputes with Ukraine (a country that transits Russian gas onwards to the EU).

The Commission’s 2002 proposal aimed to enhance measures internally within the EU to respond to the sort of gas supply crises that ensued in 2006 and 2009. The proposal ultimately led to the adoption in April 2004 of the Council Directive 2004/67/EC “concerning measures to safeguard security of natural gas supply” that emphasised the importance of infrastructure interconnections within the EU, enhanced cooperation among member states and the development of response plans for industry, member states and the EU level in case of major disruptions in supply. In annex to the Directive a “non-exhaustive list of instruments to enhance the security of gas supply” includes measures such as (among others):

Improving pipeline capacity to allow gas supplies to be diverted within the

S EU; Improving domestic production of gas;

S Diversifying sources of gas supply; and S Investing in infrastructure (pipelines and LNG terminals) for importing gas. S

237 CPI into EU external gas infrastructure policy

Generally, the measures outlined in the Directive focus on measures within the capacity and competence of the EU at that time, namely by encouraging member state solidarity in creating (physical) links in the internal energy market. A strong emphasis on subsidiarity and proportionality means that a three-step response approach was agreed under the Directive. First, industry will respond to supply interruptions with their own measures. Next, member states will step in if industry cannot respond adequately in cases of supply disruptions. Only if member states cannot sufficiently respond to the situation will the EU level step in as a third-level measure.

In 2008, the Commission reported on the implementation of the 2004 Directive (European Commission, 2008c). Here, the Commission highlighted that the most severe gas supply crisis to have taken place since the adoption of the Directive was the 2006 interruption of gas supplies. This crisis was managed effectively by measures in place at the national level, and EU-level emergency responses were not required. The crisis did mean that the EU Gas Coordination Group established by the 2004 Directive (Art. 7) met for the first time (European Commission, 2008c, p. 6). Given the shortness of the crisis (36 hours), however, and the adequacy of national measures to respond to the situation, by the time the Gas Coordination Group met, full supplies of gas had already been restored (ibid., p. 7). The difficulty in ensuring a sufficiently quick emergency response to sudden interruptions in gas supplies at the EU level indicated some inadequacy in the measures laid out in the 2004 Directive. This was further compounded by the 2009 gas crisis that was more severe.

Another important moment in the development of EU energy policy generally, and EU energy security policy more particularly, came with the 2008 second strategic energy review. The Commission released the second strategic energy review, subtitled “an EU security and solidarity action plan”, in November 2008 (European Commission, 2008e). This communication is often considered a watershed moment in the development of EU energy policy and in the security of supply policy (interviews 22 & 24; see also chapter two). It followed the 2005 informal meeting of heads of state and government at Hampton Court in the UK that emphasised increased EU coordination in energy policy, and it came after the EU agreed the 20-20-20 targets for 2020 in the March 2007 European Council (see chapter two; European Council, 2007). The second strategic energy review called especially for promoting further energy infrastructure to meet the EU’s energy needs (European Commission, 2008e, p.

238 CPI into EU external gas infrastructure policy

4). In terms of natural gas infrastructure, the review calls for special priority to supporting the Southern gas corridor initiative to bring supplies of gas from the Caspian Sea Region (and potentially also from the Middle East) to Europe, and to support projects to develop LNG capacity in the EU (ibid., pp. 4-5). One interviewee outlined that this particular communication and its acceptance by the Parliament and Council allowed for more ambitious infrastructure promotion to follow (including under the EEPR, see below, which provided an unprecedented amount of financial support to energy infrastructure projects in the EU; interview 24). In its report on the implementation of the European energy programme for recovery (EEPR, see section 6.3.3. below), the Commission readily agreed that the second strategic energy review “set the priorities for the EU in the energy field for the coming years” (European Commission, 2010f, p. 2).

In January 2009, a dispute between Russia and Ukraine caused several weeks of disruptions in gas supplies to the EU. This crisis was labelled “unprecedented” by the EU (European Commission, 2009c, p. 2; see also European Council, 2009a, p. 9). It demonstrated major weaknesses and vulnerabilities in the emergency responses of the EU in the face of such severe supply disruptions. As a result, the EU agreed to a revision of the 2004 Directive in the shape of Regulation (EU) 994/2010 concerning measures to safeguard security of gas supply and repealing Directive 2004/67/EC.

While the context of gas supply security in the EU revolves around three main issues – increasing import dependence; increasing dependence on Russia; and vulnerability to gas supply interruptions – the 2004 Directive and the 2010 Regulation respond only in a short-term way to the third challenge. In its 2008 report on the implementation of the 2004 Directive, the Commission described the challenge of defining measures for ensuring the security of gas supply in the long-term also (European Commission, 2008c, pp. 3-4). With the long-term perspective to 2050 decarbonisation, I analyse below the level of CPI in the EU’s policies to support external gas infrastructure projects.

6.3.2 Trans-European Networks for Energy (TEN-E) TEN-E policy first developed in the EU in the 1990s. It focuses on promoting the development of infrastructure projects for energy. The TEN-E guidelines were first put in place in 1996, under Decision 1254/96/EC of the European Parliament and the

239 CPI into EU external gas infrastructure policy

Council. This Decision laid down guidelines for assigning the label of a “project of common interest” to trans-European gas (and electricity) infrastructure projects, to aim for:

- “the introduction of natural gas into new regions, - the connection of isolated gas networks to the interconnected European networks, including the improvements needed to the existing networks for this purpose and the connection of the separate natural gas networks, - increasing the transmission (gas delivery pipelines), reception (LNG) and storage capacities needed to satisfy demand, and diversification of supply sources and routes for natural gas.” (Art. 2(2)).

The Treaty on the European Union (TEU, Maastricht Treaty), which entered into force in 1993, provided the legal basis for policy on trans-European networks in Title XII “Trans-European Networks”, with Articles 129b to 129d. With the Amsterdam Treaty in 1997, the wording of the articles remained largely unchanged, but they were renumbered as Title XV and Articles 154 to 156. The Nice Treaty left the Articles untouched, but with the Lisbon Treaty, these were renumbered as Title XVI and Articles 170 to 172, with “Community” being replaced with “Union” in the wording of the articles. Thus, the legal basis for action on trans-European networks for energy faced no major substantive changes throughout the period 2000 to 2010.118

In 2001, the Commission (with DG Energy and Transport in the lead: today and hereafter “DG Energy”) proposed an amendment to the 1996 TEN-E Decision (European Commission, 2001c). As stated by the Commission, the revision aimed to “underline political support” for a special category of priority TEN-E infrastructure projects of common European interest (ibid., pp. 23, 40). The revision outlined “criteria upon which support will be determined; notably the creation of the internal market, security of supply, integration of renewable energy, enlargement and the integration of ultra-peripheral regions” (ibid., p. 23). The Commission suggested gas connections directly between Russia and Germany (later realised as the Nord Stream pipeline, construction completed in 2012), connections between the Caspian Sea region and the EU (the “Southern Gas Corridor”, yet to be realised), and a gas

118 See Annex II for the full text of Title XVI, TFEU, on trans-European networks.

240 CPI into EU external gas infrastructure policy pipeline connection between Algeria and Spain (construction of the Medgaz pipeline between Algeria and Spain was completed in 2008), as the three “priority axes” of European interest to benefit from specific financial and political support in this 2001 communication (ibid., p. 25).

It may be interesting to pause here and provide some background information on two of these external infrastructure projects – namely Nord Stream and the Southern gas corridor. The North European Gas Pipeline (now known as Nord Stream) is a set of twin submarine pipelines in the Baltic Sea, connecting Vyborg in Russia to Greifswald in Germany. The project is considered to increase EU gas security by creating a direct link between the EU and Russia and bypassing transit countries. The first of the twin pipelines commenced operations in November 2011, while the construction of the second pipeline was completed in April 2012. Together, these pipelines have the capacity to carry 55bcm of gas from Russia directly to Germany every year. They have an expected lifetime of about fifty years (until beyond 2060). From its early days (discussions on the pipelines began as far back as 1993), the Nord Stream project received unofficial political support from many EU member states (Larsson, 2007). With the 2003 revision of the TEN-E guidelines, the project received official EU-level political backing as a “priority project” with opportunities for it to access financial support also. The Nord Stream consortium did not, in the end, make use of the potential funds available under TEN-E, and is completely privately funded. However, the political support attached to the “priority project” label certainly aided its realisation in terms of negotiations for financial backers and getting the go-ahead from Baltic states (interview 13).

The Southern gas corridor is an initiative that is designed and promoted to increase the EU’s security of gas supply by diversifying suppliers. Instead of importing more gas from Russia, the EU-backed pipelines in this proposed corridor (namely the Italy- Turkey-Greece Interconnector, ITGI, and Nabucco pipelines) would source gas from the Caspian Sea region, with the potential to expand later to source supplies from the Middle East. These regions have substantial reserves of natural gas. The main supplier countries proposed for the Southern gas corridor include Azerbaijan, with a proven gas reserve of 1.3tcm at the end of 2011, Turkmenistan with 13.4tcm, Iraq with 3.2tcm, and potentially Egypt (2.2tcm) and Iran (33.1tcm) in the future (political relations pending; BP, 2012, p. 20). This gas corridor would thus allow the EU access

241 CPI into EU external gas infrastructure policy to some of the largest reserves of natural gas in the world (for comparison, Russia’s gas reserves at the end of 2011 are listed as 44.4tcm – the largest reserves of any country, followed by Iran; ibid.). As of 2013, the ITGI pipeline is no longer in the race to access gas supplies from Azerbaijan (the first expected supplier for the Southern corridor with 10bcm of gas supplies per year expected to flow to the EU before 2020119), and Nabucco has had to replace its plans for a 4,000km 31bcm capacity pipeline from Austria to the Turkish/Georgian border with a shorter “Nabucco West” pipeline proposal from Austria to the Bulgarian-Turkish border (1,300km, with a capacity of between 10 and 23bcm). Nabucco West is one of two remaining pipelines bidding for the first gas supplies from Azerbaijan (the other being the Trans-Adriatic Pipeline, TAP, that has received financial support under TEN-E, but did not receive the “priority project” label). As of April 2013, the future of Nabucco remains uncertain, as decisions from Azerbaijan on allocating its supplies to one pipeline or the other have been delayed time and again.

The Commission published its proposal for a revision to the TEN-E guidelines in 2001 (European Commission, 2001c). It moved through two readings in the Parliament and Council before being adopted as Decision 1229/2003 on 26 June 2003. In neither the Parliament nor the Council did internal environmental or climate actors play a visible role in the decision-making process. In the Parliament, the ITRE committee drafted the Parliament’s first and second reading opinions, with the ENVI committee not considered for an opinion.120 In the Council, the final decision was taken in the general affairs meeting in June 2003, although the proposal had previously been discussed in the industry Council meeting in June 2002, the transport, telecommunications and energy Council meeting in November 2002 and also in the education, youth, culture and sport Council meeting in February 2003. There was no discussion in the environment Council meetings.

While the policy negotiations did lead to some changes between the proposed text and the final act, there was no change to the suggested external gas infrastructure projects (outlined in the Annexes to the Decision). Both the proposal and the final act

119 See euractiv.com for up-to-date reports on the status of the negotiations for gas supplies from Azerbaijan.

120 Two committees provided an opinion to the ITRE committee: the budgets committee (BUDG) and the economic and monetary affairs committee (ECON).

242 CPI into EU external gas infrastructure policy highlighted gas pipelines between Algeria and Spain, between Russia and the EU, and from the Caspian Sea and Middle East to the EU, as well as LNG terminals in Belgium, France, Spain, Portugal and Italy as “axes for priority projects” (Annex I). This means that projects listed under Annex III as “projects of common interest” that fall within the “axes for priority projects” listed in Annex I “shall have priority for the grant of Community financial aid…” (Art. 7).

Another revision of TEN-E guidelines amended the 2003 Decision with Decision No 1364/2006/EC. This revision was deemed necessary mainly to incorporate the newly acceded member states (or, in 2003, in preparation for the accession of these member states) into the scope of the guidelines (Decision 1364/2006/EC, Recital 1). The 2006 Decision specified that the label “project of European interest” could be applied to projects that are of a cross-border nature or have “significant impact on cross-border transmission capacity…” (Art. 8(1)). These projects would receive priority in the selection process for funding under TEN-E, but they would also receive “particular attention” under other Community funds (Art. 8(1) and 8(2)). As listed in Annex I of the 2006 Decision, several external gas infrastructure projects were assigned the label “project of European interest”, including Nord Stream, pipelines connecting Algeria with Spain and Italy, two pipelines in the Southern gas corridor: the Turkey-Greece- Italy interconnector (ITGI) and the Turkey-Austria pipeline (Nabucco, see above), and a pipeline connecting Libya to Italy (Annex I).121

The process leading to the adoption of the 2006 revision of the TEN-E was initiated in December 2003 (just six months after the adoption of the previous TEN-E Decision) with the Commission’s proposal (European Commission, 2003b). In the Parliament, the ITRE committee was responsible for providing an opinion on the Commission’s proposal, with MEP Anne Laperrouze (ALDE) as rapporteur. The committee presented its first reading report to the Parliament’s plenary in May 2005. The ENVI committee in this case gave an opinion (MEP Claude Turmes of the Greens political group was the rapporteur for opinion). Interestingly, the ENVI opinion to the rapporteur includes much criticism of the proposed long list of projects under the

121 In 2013, another revision of the TEN-E guidelines was agreed under ordinary legislative procedure – file: 2011/0300(COD). This round of revisions was carried out in the context of infrastructure priorities outlined in the Commission’s 2010 paper “Energy Infrastructure Priorities for 2020 and beyond” (European Commission, 2010e). This revision will not be discussed in this chapter, as I focus here on policy developments between 2000 and 2010.

243 CPI into EU external gas infrastructure policy guidelines, seemingly proposed without stringent and clear long-term criteria. The ENVI committee proposed deleting the article referring to “projects of European interest” based on the fact that the choice of such projects implies bypassing consultation and authorisation through local authorities (European Parliament, 2005, p. 69). This suggestion was not taken up in the ITRE committee rapporteur’s final report. Yet when it comes to supporting gas infrastructure, the ENVI opinion is positive on the role of gas in combating climate change:

“A well functioning gas market needs regulation on access to gas storage. Europe needs also to regularly improve its gas efficiency with long-term instruments such as quantitative commitment by distributors to reduce consumption. Only such measures will allow the EU to fulfil its commitment on climate change” (European Parliament, 2005, p. 60).

As this quote implies, the ENVI committee saw a continued role for gas in the EU energy mix, provided efficiency gains were realised, the internal market was upgraded and long-term measures to reduce gas consumption were adopted. Yet, there is no objection from ENVI to the support of the external gas infrastructure projects. In fact, gas is seen here as an essential component of the fight against climate change. This is understandable as a short-term solution to moving away from the more GHG- intensive use of coal. Yet the risk of “carbon lock-in” from promoting gas in the energy system was not considered in the committee procedures on the TEN-E guidelines, and not even in the ENVI committee.

In the Council, the proposal was discussed three times in the transport, telecommunications and energy Council formation in 2004 and 2005, before being agreed upon in the competitiveness Council meeting (on internal market, industry, research and space) in 2006. The Council adopted its common position on 8 December 2005, and substantially altered the Decision. It deleted the articles relating to the promotion of the “project of European interest” label and the nomination by the Commission of a European coordinator for such projects (Council of the European Union, 2005a). In the explanation of its position, Council described concern that the use of a separate category of projects would result in a “negative effect” on the realisation of other viable projects (Council of the European Union, 2005b). With regard to the appointment of a European coordinator, the Council argued that “far less

244 CPI into EU external gas infrastructure policy bureaucratic provisions could be retained for the same purpose” (ibid., p. 4). Despite informal trilogue negotiations during the first reading, these issues could not be resolved. The Commission stated, in its response, that it could not accept the text as proposed by Council (European Commission, 2005a), and the file moved to second reading in the Parliament in January 2006. The Commission especially called on Council to take note of the conclusions of the informal Hampton Court meeting of heads of state or government in October 2005, and quotes former UK Prime Minister Tony Blair, that: “in respect to energy, there was an agreement to take forward work in the energy sector, including how we try to establish a common European grid. …it is important too that energy policy is something that we work on together as a European Union…” (European Commission, 2005a, p. 5).

The second reading report adopted by the Parliament was the result of compromise negotiations between Council and Parliament that reinserted the “project of European interest” category and the “European coordinator”. This time, the Commission could accept all (17) amendments proposed by the Parliament (European Commission, 2006e, p. 3; European Parliament, 2006a). As part of the compromise, and on member state insistence, the Commission will not be able to withdraw the declaration of European interest from a project. The second reading decision by Parliament came on 4 April 2006 and Parliament and Council signed the final Decision on 6 September 2006.

6.3.2.1 CPI in the policy process of the TEN-E guidelines As outlined in chapter three, I measure the level of CPI in the policy process with reference to three main indicators, namely: the involvement of pro-climate stakeholders internally to the policymaking process; the involvement of external pro- climate stakeholders; and the recognition (and articulation) by policymakers of the functional interrelations between policies to support external gas infrastructure (in this case) and long-term climate policy objectives to decarbonise the energy sector by 2050 (see Table 3.1).

Internal pro-climate stakeholders Throughout the negotiations on the TEN-E guideline revisions, leading to both the 2003 and to the 2006 Decisions, there is limited evidence of internal environmental or climate voices playing a role. Internally to the decision-making processes in the EU,

245 CPI into EU external gas infrastructure policy

DG Energy was in the lead in the Commission, with the ITRE committee drafting the first reading opinion report in the Parliament and various formations of the Council negotiating the texts, but notably never the environment Council. In the lead up to the 2006 TEN-E Decision, the ENVI committee in the Parliament did provide an opinion on the proposed text for the consideration of the ITRE committee. Their concerns were with the label “project of European interest”, and less with long-term concerns for carbon lock-in into natural gas infrastructure. Interestingly, the only real consideration of natural gas infrastructure in the ENVI committee’s opinion to the ITRE committee was to promote efficient use of natural gas in the short-term as part of the solution to combating climate change (European Parliament, 2005, p. 60). While such a position may be understandable in the short-term as gas represents an alternative to coal, for example, the support of infrastructure represents a long-term strategy, with the infrastructure expected to operate for at least fifty years. Here, it is clear that even pro-climate actors in the EU either did not promote or could not promote long-term climate-friendly solutions that would see no support for further external gas infrastructure. It seems that the electricity interconnections received more attention from internal climate stakeholders than gas interconnections. As long as renewable energy was mentioned and highlighted in the TEN-E guidelines (which was the case), it seems the internal environmental voices were satisfied. In the case of the TEN-E revision process, pro-climate voices were not very active or visible in the policy process, and neither did they represent the long-term pro-climate position that would promote the abolition of political and financial support for fossil fuel infrastructure (in this case, external gas infrastructure). As such, internal pro-climate stakeholder involvement is very low (with little to no involvement and visibility) as even when a traditional pro-climate voice was involved it did not promote CPI (see Table 3.1).

External pro-climate stakeholders External pro-climate stakeholders were not visible nor were they involved in the policy process leading to the revisions of the TEN-E guidelines in 2003 and 2006. One interviewee reported that the main stakeholders consulted, both formally and informally, in both these policy processes were energy companies and gas and electricity network operators (interview 24). The consultations carried out by the Commission in the lead up to the TEN-E proposal published in 2001 included a

246 CPI into EU external gas infrastructure policy questionnaire that was deemed to have reached all “relevant parties” (European Commission, 2001c, p. 4). These included member states, national energy regulatory authorities, network operators, gas and electricity industries, consumers, traders and other infrastructure users, and financial institutions. Renewable energy companies may be subsumed under “electricity industries” and “other infrastructure users” in this list, but there is no indication of input from environmental NGOs, for example. For the 2006 TEN-E Decision, the Commission opened an online consultation between 25 July 2003 and 15 September 2003, which received 17 replies, with the majority of replies (13) referring to the electricity sector (European Commission, 2003a, p. 28). Five replies discussed the gas sector. These contributions came mostly from electricity transmission system operators (TSOs), European organisations representing electricity transmission and gas or oil companies, energy companies, regional and local authorities and environmental groups (European Commission, 2003b, p. 10). Only in response to one specific electricity interconnection were environmental voices raised referring to concerns over ecosystem preservation (European Commission, 2003a, 2003b). These concerns referred to a single electricity project on the Austrian/Italian border and the potential health risks and damage to the local environment of this particular project (European Commission, 2003a, p. 32).

In this case, the environmental interests were local nature conservation interests that did not question the broader environmental or climate impacts of supporting infrastructure projects. They were not focused on addressing support for external gas infrastructure. The long-term climate concerns that the promotion of gas infrastructure would risk locking the EU into a carbon system were not raised. The existence of TEN-E policy since the 1990s in the EU also meant that relations between policymakers and, especially, energy companies were well developed (Vasileiadou & Tuinstra, 2012; interview 24). It seems that external pro-climate stakeholders were not specifically targeted for consultation in TEN-E policy. Procedures for consultation were in place for the revisions, so it is not a case that such stakeholders are deliberately excluded. It is more likely that they simply are not considered to be among the most relevant stakeholders in such policy processes. It seems that pro- climate stakeholders themselves do not seek to raise the long-term climate concerns in infrastructure policy either. They do not seem to perceive this policy sector as a priority for their involvement, as they consider the greatest urgency to be a reduction

247 CPI into EU external gas infrastructure policy in coal consumption (interviews 7 & 8). This is a case of stakeholders choosing their battles (Hauser, 2011), and policymakers working with their traditional partners. The lack of external pro-climate stakeholders in the policy process means that CPI is not promoted through them – their involvement is very low CPI (see Table 3.1).

Recognition of functional interrelations The very low involvement of both internal and external pro-climate stakeholders in the TEN-E policy processes seems to be linked to the lack of any recognition or articulation in the policy process of the functional interrelations between long-term climate policy objectives and external gas infrastructure policy. As outlined above in sections 6.1 and 6.3.1, the interrelations between external gas infrastructure and combating climate change play out in the long-term horizon. While it is reasonable to expect gas to play a role in the short-term as an alternative to coal (see Figure 6.4), long-term climate policy requires a total decarbonisation of the EU’s energy sector by 2050, but with GHG emissions required to peak (globally) already around 2015.122 EU external gas infrastructure already had the capacity in 2010 to import 626.5bcm of gas per year (see Table 6.1; infrastructure capacity in the EU27 in the early 2000s stood at approximately 450-550bcm; European Commission, 2003a; BP, 2012). When gas consumption under a decarbonisation scenario needs to fall dramatically to 2050, it is clear that no further external gas infrastructure is required if decarbonisation objectives are to be achieved. The recognition of these interrelations in the policy discourse should lead policymakers to abandon political and financial support of external gas infrastructure. As described above, this was certainly not the case for the TEN-E revisions. Not only were these long-term climate policy objectives unrecognised in the policy process by policymakers, but also even the traditional pro- climate voices did not highlight these interrelations. Where climate policy objectives were mentioned, they were seen to benefit from further gas infrastructure as a short- term climate solution. With these infrastructure projects taking many years to be realised, and having lifetimes of fifty years or more, it is clear that the long-term functional interrelations were not recognised. Thus, this demonstrates no or very low CPI (see Table 3.1).

122 With GHG emissions needed to peak around 2015, it makes little sense to add new fossil fuel capacity of any sort (coal, oil or gas) to the energy system.

248 CPI into EU external gas infrastructure policy

In summary, the level of CPI in the policy process leading to the revisions of the TEN-E guidelines is very low or even non-existent (see Table 6.3). This is an aggregate of the very low or no levels of CPI evident through the role of internal and external pro-climate stakeholders in the policy process and the lack of recognition of functional interrelations between external gas infrastructure policy and long-term climate policy objectives.

6.3.2.2 CPI in the policy output of the TEN-E guidelines The discussion in section 6.1.3 demonstrates that there is no need for further external gas infrastructure under decarbonisation scenarios, and thus no need for policy to support such infrastructure. An abandonment of such support policies would represent the highest level of CPI in policy output that can be expected (section 6.2). It is difficult to conceive of a policy output that provides political and financial support to several external gas infrastructure projects as demonstrating any levels of CPI. The rules governing the financial aid under TEN-E are laid down in Council Regulation (EC) No 2236/95 on the “granting of Community financial aid in the field of trans- European networks”. As stipulated in Article 4, aid for TEN-E projects can include co-financing of up to fifty per cent of the cost of preparatory, feasibility and evaluation studies, subsidising the interest on loans from the European Investment Bank or other financial entities, contributing towards fees to guarantee loans, and providing direct grants in certain cases. Generally, the budget of the TEN-E programme has been relatively modest, with an annual envelope of about EUR 20-23 million (European Commission, 2003a, p. 9; interview 24), which, in 2003, was reported to allow funding for between ten and twenty projects, with the funds being used mainly for the initial study phases of the projects (European Commission, 2003a, p. 11). The political importance of the labels for projects assigned under the TEN-E guidelines, however, should not be underestimated. Even in cases where no funding has been applied to a project (such as with Nord Stream), the label “project of European interest” has helped in assigning credibility and status to the project (interview 13). Indeed, the Commission outlines that the indirect effects of the recognition of a project as one of “European interest” may by more important for the project’s development and advancement than any financial input (European Commission, 2003a, p. 11).

249 CPI into EU external gas infrastructure policy

Political and financial support for energy infrastructure projects in general need not be in conflict with climate policy objectives, if those projects are linked to ensuring climate policy objectives are met (such as upgrading the electricity grid to handle increased renewable generation). The very existence of external gas infrastructure projects in the TEN-E guidelines, however, suggests that this part of the policy is in conflict with climate policy objectives. As such, CPI in the policy output of TEN-E guidelines on external gas infrastructure can only be measured as none or very low (see Table 3.2). The projects supported in the 2003 and 2006 TEN-E guidelines have had mixed levels of success. In terms of gas pipelines, since the agreement of the 2003 guidelines (and as of April 2013), construction has been completed in the 8bcm Medgaz pipeline (2008) and the 55bcm Nord Stream pipeline (2012). The supported Southern gas corridor still has no guaranteed sources of gas (up to 23bcm), the Galsi pipeline (8bcm) has not yet begun construction, and the ITGI pipeline has been excluded by Azerbaijan as a candidate for its supplies of gas. The infrastructure projects constructed (or planned) since 2000 will remain in place beyond 2050, when the EU should have already achieved its long-term decarbonisation goals. In this case, the TEN-E guidelines actually help widen the gap between the status quo and very high or complete levels of CPI (see also the discussion on measuring CPI in the policy output in chapter three).

6.3.3 European Energy Programme for Recovery (EEPR) As part of the EU’s response to the 2008 economic crisis, a European energy programme for recovery (EEPR) was agreed in 2009. This first Regulation (No 663/2009) outlined the conditions and eligibility requirements for infrastructure projects related to gas, electricity, CCS and offshore wind to access funding directly from the EU. Additionally, eligible projects were listed in the Regulation’s Annex. In terms of external gas infrastructure, the Regulation earmarks funding for four external gas infrastructure projects (out of a total of 18 gas infrastructure projects amounting to EUR 1.44 billion):

EUR 200 million for the Nabucco pipeline project in the Southern gas

S corridor; EUR 100 million for the Italy-Turkey-Greece Interconnector (ITGI) –

S Poseidon pipeline project (also in the Southern gas corridor);

250 CPI into EU external gas infrastructure policy

EUR 120 million for the Galsi pipeline connecting Algeria and Italy;

S And EUR 80 million for an LNG terminal on the Polish coast (Regulation S (EC) No 663/2009, Annex).

The funding is awarded to project promoters as grant funding, and can cover up to fifty per cent of the eligible investment costs for gas projects (as outlined in Art. 9). Together, these four projects are expected to supply the EU with an additional 50bcm of natural gas supplies per year (European Commission, 2010f, p. 4), although their combined maximum capacity would be about 57bcm.123

The EEPR deals with long-term natural gas supply decisions and applies concrete policy support (in the form of promises of funding) to certain external gas infrastructure projects. The motivations for this policy measure include ensuring such costly projects advance even under an economic context that discourages investment (under the financial and economic crises that began in 2008); and to ensure infrastructure is in place to meet the long-term goals of the security of gas supplies, including the diversification of gas importing routes and sources (as well as upgrading the internal EU grid to ensure gas flows throughout the Union). Article four of the Regulation lists its specific objectives, including enhancing the EU’s security of energy supply through the “diversification of sources of energy, routes and supplies” (Art. 4(a)).

The 2009 Regulation was proposed and adopted under Article 156 (on trans-European networks, now Articles 170-172 TFEU) and Article 175(1) of the environment chapter of the Treaty establishing the European Community (TEC).124 In the policy process leading to the adoption of the EEPR Regulation, it is particularly interesting to note that the Commission did not open a public consultation, nor did it prepare an impact assessment. This was justified due to the urgent need to respond to the economic crisis (and the fact that funds needed to be committed within the year), and hence the lack of time to complete a consultation process and develop an impact assessment (European Commission, 2009d, pp. 2-3). In the Parliament, MEP

123 Considering the objective of the EEPR was to boost the economy and create jobs within the EU, questions could also have been raised about whether big external gas infrastructure projects met these objectives.

124 See Annex II for the text of Article 175(1), TEC and of Articles 170-172, TFEU.

251 CPI into EU external gas infrastructure policy

Eugenijus Maldeikis (Union for Europe of the Nations political group) served as rapporteur in the industry, research and energy committee (ITRE). The environment committee (ENVI) decided not to give an opinion. In the Council, the proposal was discussed once in the energy Council, twice in the general affairs Council and twice (including for the final Council agreement) in the economic and financial affairs Council meeting. From 28 January 2009, when the Commission presented its proposal, less than six months passed before the final act was signed on 13 July 2009.

In its April 2010 review of the implementation of the 2009 Regulation, the Commission expressed a general satisfaction with the implementation of the EEPR (European Commission, 2010f). The implementation procedure began with a call for proposals in the three sub-programmes of the EEPR (gas and electricity infrastructure, offshore wind and CCS) in May 2009, with 33 proposals for gas infrastructure projects submitted. After evaluation by the Commission and relevant committees (principally, the TEN-E financial assistance committee), the Commission decided to grant funding to 31 of the gas projects for a total of EUR 1.391 billion (European Commission, 2010f, p. 7). The Commission described how the EEPR, at this early stage in its implementation, seemed to be acting as a stimulus for attracting investors to the supported projects (ibid., p. 8).

By 31 May 2010, the Commission proposed an amendment to the 2009 EEPR Regulation to assign the uncommitted funds – estimated at this time to be EUR 114 million, provided none of the projects fail (European Commission, 2010h, p. 2), but the final fund amounted to EUR 146 million (European Commission, 2011i, p. 10; 2012b, p. 2)125 – to support energy efficiency and renewable energy initiatives. This proposal came in fulfilment of Recital seven of the 2009 Regulation that states: “in the event that it is not possible to commit all funds by the end of 2010, the Commission has declared its intention, if appropriate, to propose, when reporting in 2010 on the implementation of this Regulation, measures allowing for the financing of projects consistent with the Recovery Plan, such as projects in the areas of energy efficiency and energy from renewable sources”. The ensuing 2010 Regulation (No 1233/2010) created the European energy efficiency fund, adopted under the legal

125 Further funds for the energy efficiency fund were added by European Investment Bank, Deutsche Bank and the Italian investment body Cassa Depositi e Prestiti, bringing the total funds available to EUR 265 million (ENDS Europe, 2011b).

252 CPI into EU external gas infrastructure policy competence of the new energy chapter in the TFEU (Art. 194(1)(c) on promoting energy efficiency and renewable energy).126 Again, the decision-making process was speedy, from the presentation of the Commission’s proposal on 31 May 2010, through first reading agreement in the Parliament and the Council to the signature of the act on 15 December 2010. The responsible committee in Parliament was the ITRE committee, with Kathleen van Brempt from the Progressive Alliance of Socialists and Democrats political group as rapporteur. In the Council, final agreement was made in the employment, social policy, health and consumer affairs Council meeting on 6 December.

I focus here on the elements of the 2009 Regulation that provide financial (and political) support for external gas infrastructure projects.

6.3.3.1 CPI in the policy process of the 2009 EEPR Regulation

Internal pro-climate stakeholders With the involvement of internal pro-climate stakeholders (such as DG Environment, the ENVI committee in the Parliament, and the environmental Council) in the internal decision-making processes, there are more opportunities for CPI to be advanced. In the case of the 2009 EEPR Regulation, however, these pro-climate stakeholders were missing from the policy process. In the Commission, DG Energy was in the lead on the policy proposal, and while they proposed support for offshore wind projects as well as the external gas infrastructure projects discussed above, at no point did any internal pro-climate stakeholder question the inclusion of gas infrastructure projects. There was no impact assessment procedure for the EEPR regulation, which may also have limited cross-sectoral discussions on the Regulation within the Commission. In the Parliament, the ENVI committee decided not to give an opinion on the policy proposal, and in the Council, the proposal was not discussed in the environment Council formation. Furthermore, the speed of the decision-making process (less than six months) meant that little change was made to the Commission’s original proposal. As 2009 was an election year in the Parliament, and as the proposal was sped through, the EEPR was not even discussed or negotiated in detail within the ITRE committee, let alone among different committees (interviews 25 & 26).

126 See Annex II for the text of Article 194(1)(c), TFEU.

253 CPI into EU external gas infrastructure policy

Under such circumstances, the pro-climate stakeholders in the EU decision-making processes were not involved and were not visible in the policy process leading to the 2009 EEPR Regulation. Thus, the level for this indicator is none or very low (see Table 3.1).

External pro-climate stakeholders The 2009 EEPR Regulation was proposed and negotiated under unusual circumstances for the co-decision procedure. It was proposed by the Commission as a quick response to the economic and financial crises. The Commission thus did not carry out the usual consultation procedures required under policies proposed under the co-decision procedure. There was no open consultation process, and neither was there an impact assessment accompanying the proposal (European Commission, 2009d). Thus, unlike in the case of the TEN-E revisions, and unlike the other cases discussed in this dissertation, the usual procedures for consultation were not open to external stakeholders. The access of stakeholders was thus limited to later stages in the negotiations after the Commission had made its proposal, and to informal lobbying activities with the Parliament and Council.

Even so, pro-climate stakeholders were not visible and neither were they involved in the policy process leading to the 2009 EEPR Regulation. As with the TEN-E Decisions, this could be due to the lack of recognition of the functional interrelations between long-term climate policy objectives and external gas infrastructure (see below); or it may be related to the lack of pro-climate stakeholder resources to get involved in every policy process. Beyond the limited access, especially in the early stages of the policy process, the speed of the negotiations in this case also left many pro-climate stakeholders outside the process (interviews 8 & 24). With Parliament insisting on including a Recital in the Regulation that unspent funds be used for energy efficiency and renewable energy projects (creating the energy efficiency fund and resulting in the 2010 EEPR Regulation, see above), external pro-climate stakeholders acted mostly to ensure that this obligation was fulfilled later. In the process leading to the 2009 EEPR they were neither involved nor visible in the process to the extent that they could raise levels of CPI in objection to the external gas infrastructure support provided by the EEPR. The level for this indicator is none or very low (see Table 3.1).

254 CPI into EU external gas infrastructure policy

Recognition of functional interrelations

As with the TEN-E revisions, discussed above, it is plausible that the low involvement of internal and external pro-climate stakeholders is related to a lack of recognition (by all parties) of the functional interrelations between promoting external gas infrastructure and achieving long-term climate policy objectives. Where climate change was mentioned in the policy process, it was rather in reference to the other projects supported under the EEPR (namely, CCS and offshore wind projects), as they are seen as responding directly to the climate challenge. However, even within this one policy measure, there was no explicit questioning of the conflict between supporting external gas infrastructure projects while at the same time trying to achieve long-term climate policy objectives and decarbonisation. Even within a single policy measure, conflicts are present. Recognition of the functional interrelations could rather have led to a questioning of the need for further support for such infrastructure projects by policymakers and even a reduction in the amounts of funding assigned to them. At best, external gas infrastructure projects could have been excluded from the EEPR. As the policy process unfolded, the Galsi pipeline was added to the original proposal, and external gas infrastructure projects were earmarked for EUR 500 million (compared to EUR 430 million in the Commission’s proposal; European Commission, 2009d, p. 25). In the case of the external gas infrastructure support provided by the EEPR Regulation, there is no evidence of the recognition of functional interrelations: a none or very low score for this indicator (see Tables 3.1 and 6.3).

6.3.3.2 CPI in the policy output of the 2009 EEPR Regulation In many respects the EEPR Regulation builds on the policy measures of the TEN-E framework. The external gas infrastructure projects identified under EEPR had already received preliminary funding under TEN-E for their initial scoping investigations and/or are considered as “projects of European interest” under TEN-E. While TEN-E provides support for projects in their early stages of development, EEPR targets more mature projects that are likely to need funding support to develop further under the context of the economic downturn. Additionally, the funding earmarked under TEN-E is limited. TEN-E makes available an annual financial envelope of about EUR 20 million (European Commission, 2010f, p. 3), whereas the EEPR earmarked EUR 500 million for the four external gas infrastructure projects

255 CPI into EU external gas infrastructure policy listed above, and a total of EUR 1.44 billion for all gas infrastructure projects (of which EUR 1.391 billion was allotted to various projects in the end; European Commission, 2010f, p. 7; Regulation (EC) No 663/2009 Annex A.1).127

As outlined above in section 6.2, very high levels of CPI would indicate that no support (financial or political) should be given to external gas infrastructure projects that are not needed under decarbonisation (see Table 6.2). While overall the EEPR also aimed to support electricity infrastructure projects, offshore wind infrastructure projects and CCS projects, I focus on the support for the four external gas infrastructure projects. At the time of the agreement on the EEPR, the Nord Stream pipeline was under construction and no longer required the political support provided by the TEN-E Decisions. Nord Stream creates a further 55bcm of external gas infrastructure capacity (the twin pipelines were completed in 2012). The four projects supported by EEPR would add a maximum of a further 57bcm capacity of external gas infrastructure. These projects seem to have been inherited from the TEN-E Decisions, but the financial support made available under EEPR is substantially larger than anything available under TEN-E. The EEPR not only continues the misplaced political and financial support of external gas infrastructure projects that are not required under long-term climate policy objectives to decarbonise by 2050, but also actually increases the financial support. Thus, the EEPR widens the gap between very high levels of CPI in external gas infrastructure policy and the policy reality. CPI in the policy output of the EEPR policy on external gas infrastructure is thus none or very low (see Table 6.3).

6.4 Explaining CPI in EU external gas infrastructure policy The case of EU policy supporting external gas infrastructure is particularly interesting for understanding CPI in the EU’s energy sector. External gas infrastructure responds first and foremost to security of supply objectives. However, energy security is just one of the three main objectives of EU energy policy (along with competitiveness and sustainability, see also chapter two). To ensure the energy security and climate change objectives of the EU are in harmony, it would be sensible to support more domestic

127 In a report in April 2011 (European Commission, 2011i), the Commission revealed the final figures committed to gas infrastructure projects to be EUR 1.363 billion. While the Polish LNG terminal was under construction at this point, Nabucco, Galsi and ITGI all faced delays. By August 2012, still no payments had been made to Nabucco or Galsi while six per cent and 30 per cent of the promised funds had been made to the ITGI-Poseidon project and the Polish LNG terminal, respectively (European 2012b, pp. 3-4, 6).

256 CPI into EU external gas infrastructure policy sources of renewable energy (see chapter four) and improve energy efficiency measures (see chapter five) to reduce energy consumption overall. Together, such policy measures can respond to climate and energy security objectives. However, policies such as the TEN-E guidelines and EEPR do not display harmonious policy objectives when it comes to the political and financial support of external gas infrastructure. What can explain the very low and nearly non-existent levels of CPI in the policy output and process of the TEN-E guidelines and the EEPR Regulation (see Tables 6.3)?

CPI in the 2003 & 2006 CPI in the 2009 EEPR

TEN-E Decisions Regulation CPI in policy output None or very low None or very low Internal pro- None/very climate Very low low stakeholders: CPI in External pro- the None/very climate Very low None or None or policy low stakeholders: very low very low process Recognition None/very None/very of functional low low interrelations: Table 6.3: Levels of CPI in the policy process and output of the external gas infrastructure policy of the 2003 and 2006 TEN-E Decisions and the 2009 EEPR Regulation.

In this section I analyse the results found with respect to the four explanatory variables described in chapter three, namely: functional interrelations; political commitment; institutional and policy context (each of which explain CPI both in the process and in the output); and the process dimension (to explain CPI in the policy output).

6.4.1 Functional interrelations The functional interrelations between external gas infrastructure support and long- term climate policy objectives to reduce GHG emissions in the EU by 2050 are the same for both the TEN-E and EEPR policy measures. Where both policy measures seek to support, financially and politically, the development of external gas infrastructure in the EU, these measures can only be considered as indirectly related to

257 CPI into EU external gas infrastructure policy

(i.e. they are not closely linked to or have direct, immediate effects) and in conflict with long-term climate policy objectives. While natural gas has often been hailed as the cleanest fossil fuel, its continued use in the long-term threatens the achievement of long-term climate policy objectives (gas becomes part of the climate problem). Policies to support the expansion of external gas infrastructure risk locking in fossil fuel infrastructure, leading to risks of increased (or even continued high levels of) consumption of gas, as supplies of gas continue to be available (or even higher supplies of gas become available). The nature and type of functional interrelations between policies supporting external gas infrastructure and long-term climate policies can help explain the levels of CPI found in the policy and output, above (see also Table 3.4).

The functional interrelations are indirect, in that the construction of external gas infrastructure in itself does not necessarily lead to increases in GHG emissions. It is the risk of carbon lock-in that is at stake here. Increased capacity in external gas infrastructure means, at best, that investments are diverted away from more climate- friendly energy projects (such as renewable energy projects, energy efficiency improvements etc.). At worst, it means the EU’s energy system is locked-in to a fossil fuel infrastructure that stunts moves towards decarbonisation (see also Dupont & Oberthür, 2012). Since these functional interrelations between the policies are indirect, they may thus be obscured or hidden in the policy process. Long-term objectives may be hidden throughout the policy discourse, especially as short-term benefits of switching from coal to gas, and the short-term role of gas in combating climate change, are highlighted (European Parliament, 2005).

Additionally, the functional interrelations are conflictual. This means that achieving the objectives of policies that support external gas infrastructure development cannot be in harmony with long-term climate policy objectives to decarbonise the EU’s energy sector by 2050. As outlined above in section 6.2, no further external gas infrastructure is required to 2050 under decarbonisation scenarios. In fact, as outlined in Table 6.2, it is clear that external gas infrastructure capacity in 2010 already exceeds by more than 200bcm the actual imports of gas in that year (and also exceeds the gas consumption of the EU in 2010 by over 130bcm). Supporting further external gas infrastructure is in conflict with decarbonisation aims in that it promotes

258 CPI into EU external gas infrastructure policy continued EU energy sector development along a fossil-fuel path rather than making the necessary shift to climate-friendly and zero-GHG emissions energy production.

Such an indirect and conflictual nature of the functional interrelations between external gas infrastructure policy under the 2003 and 2006 TEN-E guidelines and the 2009 EEPR Regulation can affect the level of CPI in both the policy process and policy output hiding these interrelations in the policy discourse. When functional interrelations are indirect (and thus unclear) and conflictual, it may be more difficult for stakeholders and policymakers to see (and then articulate) the interrelations between the policy objectives.128 The nature of the interrelations in this case is unfavourable to the promotion of CPI. Thus, the interrelations are somewhat hidden and do not promote CPI either in the policy process or in the policy output.

6.4.2 Political commitment First, I discuss overarching EU commitment to combating climate change during the 2000s. EU commitment to combating climate change generally evolved over the course of time, also along with EU leadership ambitions internationally. In the lead-up to 2006, EU political commitment to combating climate change can be described as rather medium (see Table 3.5). In the early- to mid-2000s, the EU repeatedly described its ambitions to be a leader internationally in the fight against climate change, yet its internal policies did not demonstrate the concrete expression of these ambitions. This period up to 2006 can be described as an example of “rhetorical leadership” on climate change, with evidence of political commitment in statements, but not in the agreement of domestic policy although the record was beginning to improve towards the mid-2000s (Oberthür & Roche Kelly, 2008, p. 39). This overarching medium level of political commitment to combating climate change generally may be relevant for understanding the very low levels of CPI in the policy process and the policy output of the TEN-E guidelines on external gas infrastructure. Insufficient levels of political commitment generally may have resulted in climate change not appearing on the agenda of some policy sectors, for example.

The 2009 EEPR Regulation was proposed and adopted within a short six months, from proposal in January 2009 to adoption in July 2009. This measure was in direct

128 As discussed in chapter three, the more synergistic and direct the functional interrelations between two policy areas, the more likely that CPI can be advanced in the policy process and output.

259 CPI into EU external gas infrastructure policy response to the economic crisis that began in 2008. However, the negotiations on the policy took place in the wake of the agreement on the climate and energy package (December 2008) and in the run up to the international climate negotiations in Copenhagen (COP15 in December 2009). Political commitment to climate change generally in the EU was high at this time (although it began to wane in the aftermath of the international negotiations in Copenhagen; see, e.g., Falkner, Stephan & Vogler, 2010; Oberthür & Roche Kelly, 2008). This overarching high score is reflected in the concrete translation of statements of political commitment into policy measures (through the climate and energy package), combined with a push from the EU to influence the international negotiations, especially at Copenhagen (see Table 3.5).

Secondly, I assess political commitment to CPI leading to the agreement on the TEN- E guideline revisions and the 2009 EEPR Regulation. Unsurprisingly, with the lack of the recognition of the functional interrelations between long climate policy objectives and the objectives of external gas infrastructure support policy, there is no evidence of political commitment to CPI in the political statements or the subsequent policy output in the TEN-E revisions. The very suggestion of labelling external gas infrastructure projects as having “priority” and as being “projects of European interest” suggests that the political commitment in this case rather supports the expansion and continuation of policies for developing external gas infrastructure (contrary to what could be considered supporting CPI, which would be reducing and eliminating such policies; see section 6.2 above). When it comes to external gas infrastructure policies, the overarching concern is energy security and climate concerns rarely enter the picture. Indeed, among the policymakers involved in agreeing the TEN-E revisions, energy security has often been a priority goal. In the ITRE committee in the Parliament, for example, energy security has often been weighted by the majority of members as the most important energy policy objective, above competitiveness and sustainability, reducing opportunities for commitment to the promotion of CPI (interviews 25 & 26). In the Council, neither climate change nor promoting CPI seemed to enter discussions on TEN-E. Even though some mention of the need to support renewable energy appear in the Council’s common position, this is in reference mainly to the electricity infrastructure projects, and not in question to the support of external gas infrastructure (Council of the European Union, 2005b). The commitment to energy security overrides commitment to serious promotion of CPI in

260 CPI into EU external gas infrastructure policy external gas infrastructure policy under the 2003 and 2006 TEN-E guidelines. At the time the 2003 and 2006 TEN-E guidelines were being revised, political commitment to CPI within the external gas infrastructure policy under TEN-E can only be considered none or low (on the threefold scale shown in Table 3.5 with no evidence of commitment to CPI in statements).

With respect to political commitment to CPI in the external gas infrastructure policy of the 2009 EEPR Regulation, there is no evidence of commitment to CPI (none or low political commitment, see Table 3.5). Despite the overarching context of high commitment to combating climate change, it was rather the economic crisis and the ensuing poor investment context for large infrastructure projects that motivated the EEPR Regulation. With regard to external gas infrastructure in particular, the 2006 and the January 2009 gas crises helped place energy security at the forefront of considerations on infrastructure projects. The Southern gas corridor, for example, received promises of EUR 300 million for two proposed pipelines, with the major aim of sourcing natural gas from the Caspian region (and reducing dependence on Russia).129 In discussions on this and other projects, there is no commitment to the advancement of CPI. Thus, the political commitment in the case of external gas infrastructure in the 2009 EEPR Regulation was rather towards security of supply measures, and CPI did not feature.

6.4.3 Institutional and policy context At the time of the discussions leading to the 2003 and 2006 TEN-E revisions and the 2009 EEPR Regulation, several elements of the surrounding institutional and policy context may help explain the very low levels of CPI found in the policy output and process (see Table 6.3).

First, the institutional context at the time of the negotiations may shed some light on why such levels of CPI were found. Both the TEN-E guidelines and the 2009 EEPR Regulation were agreed under the co-decision procedure, with Articles 154 to 156 on trans-European networks (TEC, today Articles 170 to 172 of the TFEU) providing the

129 This particular project has received explicit political backing from the EU – including for example, from the Commission’s President José Barroso when he and his energy commissioner, Günther Oettinger, carried out negotiations with Azerbaijan to facilitate access to natural gas supplies there, in exchange for visa facilitation for Azeri nationals (See “Barroso tops Azeri gas deal with visa facilitation”, Euractiv.com, 14 January 2011, http://www.euractiv.com/energy/barroso-tops-azeri-gas-deal-visa- news-501255, accessed: 10 September 2011).

261 CPI into EU external gas infrastructure policy legal basis. The 2009 EEPR Regulation was also adopted with Article 175(1) on the environment as a second legal basis. 130 The co-decision procedure means that decisions are made under qualified majority voting (QMV) in the Council, meaning no one member state could veto a decision. In theory (see also chapters two and three), QMV may open opportunities for policy change. In the case of external gas infrastructure policy under the TEN-E guidelines and the EEPR Regulation, this does not seem to have had such an effect. It has been remarked that it can be difficult for policymakers to dramatically alter the proposal presented by the Commission (Tanasescu, 2009). This is rather the case for the TEN-E revisions and the EEPR Regulation. In particular, when the Council wished to remove the label “project of European interest” in negotiations leading to the 2006 TEN-E Decision, the Commission responded that it could not accept the omission of the label by the Council, and the proposal moved into second reading (European Commission, 2005a). The Council eventually accepted after further negotiations. In general, co-decision and QMV decision rules could be considered more favourable to advancing CPI in the policy process and output. However, interestingly for the case of external gas infrastructure policy, unanimity decision-making rules may have been more favourable for the advance of CPI as, in this case, advancing CPI would require abandoning or weakening the policy measure. With unanimity, in theory, one member state would be sufficient to block the decision. However, in neither the TEN-E nor the EEPR policy measures was unanimity in place and neither did the QMV decision- making rule favour CPI. It cannot be said that QMV was sufficient for heightened possibilities for policy change in this case, and indeed, it could be said that QMV, even in theory, is more unfavourable to advancing CPI in a case where CPI requires the abandonment of the policy measure.

In terms of the institutional context, there are certain singularities about the 2009 EEPR Regulation. The co-decision procedure usually calls for open consultations with stakeholders, but in this case no consultations took place in the name of ensuring a speedy process. Additionally, and also unusually for the co-decision procedure, the Commission prepared no impact assessment on its proposal, also due to the expressed urgency of agreeing on the Regulation. This may certainly have affected the level of

130 See Annex II for the text of Article 175(1), TEC.

262 CPI into EU external gas infrastructure policy

CPI in the process (and consequently in the output). Finally, the speed of the negotiation process (less than six months from proposal to adoption) meant that in- depth negotiations among the policymakers did not really take place. These peculiarities are rather unfavourable to the advancement of CPI in this case.

A further institutional background factor in this case is the EU enlargement of 2004. This enlargement was the stated reason for proposing a new revision to the TEN-E guidelines in 2003 – just six months after the previous revision had been adopted (European Commission, 2003b). In terms of external gas infrastructure support, the new member states from mostly central and Eastern Europe were unlikely to oppose the continuation of policy support to such projects. Many of these countries are almost entirely reliant on Russia for supplies of natural gas (although their levels of gas consumption may be lower than many other EU member states), and supporting infrastructure that reduces vulnerability due to this situation is an energy security priority (see section 6.3.1 above). This context is rather more unfavourable to CPI in policies after 2004, as the heightened emphasis on security of supplies of fossil fuels is likely to have precluded any possibility of climate concerns entering discussions.

Finally, the TEN-E guidelines have quite a history in the EU. With the Maastricht Treaty, trans-European networks became a part of the Treaty system. The first Decision on TEN-E guidelines was agreed in 1996 (Council Decision 96/391/EC), and subsequent revisions followed the basic objectives set out both in the Treaty, and in this first Decision. Article 1 of the 1996 Decision outlines the objective of “creating a more favourable context for the realisation of projects of common interest in connection with trans-European energy networks…”, but nowhere are climate or environmental objectives highlighted in this early Decision. Subsequent policy revisions follow this same prioritisation for infrastructure projects, without emphasising the environment or climate. The policy proposals and decisions on the 2009 EEPR also followed the same frame as the TEN-E guidelines. Many of the suggested projects were the same as those supported by TEN-E (except with much larger sums committed under the EEPR Regulation). As questions were not raised in relation to previous revisions of TEN-E that included support for external gas infrastructure, they were not raised for the EEPR Regulation either. Such a path dependency of past policy decisions provides the framework for future policy revisions that allows little room for dramatic policy change (Jordan & Lenschow,

263 CPI into EU external gas infrastructure policy

2010; Pierson, 1998).131 This form of path dependency means that the EEPR simply took one step further along the “path” already blazed for many years by TEN-E, with later TEN-E revisions also always following the same frame. This proved unfavourable to advancing CPI and ensuring radical policy change.

Finally, the 2009 EEPR Regulation faced a peculiar institutional context in that it was proposed and negotiated during an election year for the Parliament. It is telling that, although the ITRE committee in the Parliament was in the lead for negotiations on the proposal within the Parliament, members of that committee mentioned that they did not know the proposal or even recall negotiating it in the committee (interviews 25 & 26). Of course, this cannot be the case for all members of the ITRE committee, but it is striking that both the speed of the legislative procedure and the distraction of the upcoming elections meant little general attention was given to the proposal (making CPI even less likely to be advanced). This is a context that may be more unfavourable to advancing CPI.

The broader policy context surrounding the negotiations leading to the revisions of the TEN-E guidelines in 2003 and 2006 and the 2009 EEPR Regulation includes the fact that EU natural gas production was on a steady decline, while dependence on imports of gas was increasing, throughout the 2000s. As discussed in section 6.1.2, EU domestic natural gas production stood at 232bcm in 2000, but this figure had fallen to 212bcm by 2005, although consumption rose over the same period from 440bcm in 2000 to 494bcm in 2005 (BP, 2011, pp. 22-23). This reality, along with high dependence on mainly Russia, was not lost on EU policymakers, who discussed the importance of security of gas supply also in the negotiations on the 2004 Directive on measures to safeguard gas supply (Directive 2004/67/EC). The disruptions in gas supply to the EU in 2006 and (more importantly) in January 2009 pushed the issue of energy security of gas supplies even higher up the EU’s political agenda in the run up to the agreement on the 2009 EEPR Regulation. Such realities and such a priority on energy security measures seemed to overshadow any concerns for climate change and the risk of carbon lock-in. While energy security objectives could be pursued through energy efficiency measures and increases in domestic renewable energy generation,

131 The negotiations leading to the 2003 TEN-E revisions also took place alongside negotiations on the second internal energy market package (Directive 2003/55/EC on natural gas and Directive 2003/54/EC on electricity), which is a main motivation behind TEN-E guidelines, yet not so relevant for the specific support for external gas infrastructure.

264 CPI into EU external gas infrastructure policy the security framing in these cases was unfavourable to advances in CPI. The emphasis was on securing more infrastructure connections with current and potential gas suppliers.

With regard to the wider policy context under which the external gas infrastructure policy support of the 2009 EEPR Regulation was negotiated, the economic and financial crises that began in 2008 provided further context. These crises were already beginning to shift attention away from the urgency of combating climate change and ensuring CPI in sectoral policymaking in early 2009 (see Stern, 2007, on the economics of climate change). It was in direct response to these crises that the EU agreed the 2009 EEPR Regulation – as a way of stimulating an investment climate for large energy projects (although the job creation benefits for EU citizens of external gas infrastructure projects, for example, could have been disputed). This context was also unfavourable to advancing CPI in the policy process and output.

Finally, a particular context in the 2000s also relates to the unconventional or shale gas boom in the US that has resulted in more gas supplies (through, e.g., LNG) being available for the EU. This is linked with the construction of several new LNG terminals in EU territory and the expansion or planned expansion of existing terminals (Gas LNG Europe, 2011). LNG supplies originally destined for the US arrived instead in the EU as demand for imports of gas in the US has dropped dramatically in line with their expansion of their domestic gas production. Such a situation of ease of supply of gas and increased supply of LNG to the EU is also unfavourable to CPI, as gas supply has begot gas infrastructure (with the example of the Polish LNG terminal supported by both EU policies).

Other external events, such as the ratification by the EU of the Kyoto Protocol (2002), negotiations with Russia on the Protocol and its final entry into force (in 2005), and the international climate negotiations agenda throughout the 2000s, did not seem to impact levels of CPI in the policy output and process of external gas infrastructure policies under the 2003 and 2006 TEN-E revisions. These were not part of the policy discourse.

Overall, the institutional and policy context is rather unfavourable for policy change in favour of CPI occurring throughout the 2000s. The external context favoured a prioritisation of energy security objectives over climate concerns, with the 2008

265 CPI into EU external gas infrastructure policy economic crisis shifting attention away from climate change. The internal institutional context included enlargement with states highly dependent on a single gas supplier, past policy decisions framing the future policy development and decision-making procedures that, in this case (likely due to the lack of recognition of the functional interrelations), did not favour policy change for CPI.

6.4.4 Process dimension In the case of TEN-E revisions and the 2009 EEPR Regulation on external gas infrastructure policy, the level of CPI in the policy process was measured (see Table 6.3) as very low (or even non-existent). This score was an aggregate of the very low or no involvement or visibility of both internal and external pro-climate stakeholders and the lack of recognition of the functional interrelations between external gas infrastructure policy and long-term climate policy objectives in the policy discourse.

As an explanatory variable, the process dimension can aid in explaining levels of CPI in the policy output only. CPI was also measured to be very low in the policy output of both the TEN-E guidelines and the EEPR Regulation. In other words, the policy did not help close the gap between the status quo of policy objectives and the required policy measures to achieve long-term climate policy objectives (in this case, requiring no new political or financial support for external gas infrastructure projects that risk locking EU infrastructure into a fossil-fuel energy system). The level of CPI in the policy process is expected to affect the level in the policy output – without evidence of CPI in the policy process, it would be surprising to suddenly find high levels of CPI in the policy output (although not inconceivable; see also chapter three).

While each of the three indicators in the measurement of CPI in the policy process play a role in helping to explain CPI in the policy output, the lack of the recognition of functional interrelations between external gas infrastructure policy and long-term climate policy objectives seems crucial. In theory, environmental and pro-climate stakeholders could have seized on opportunities to involve themselves in the policy process, with online consultation procedures and informal lobbying activities in the lead up to the adoption of the TEN-E Decisions. However, even when one pro-climate internal stakeholder was involved to an extent (the ENVI committee in the Parliament during the negotiations on the 2006 TEN-E revisions), they did not express concern that the external gas infrastructure support would be in conflict with long-term climate

266 CPI into EU external gas infrastructure policy policy objectives. In this respect, even the pro-climate stakeholders did not recognise (or articulate or act upon) these functional interrelations.

The EEPR policy process was somewhat particular, in that consultation opportunities for both external and internal pro-climate stakeholders was limited due to the short timeframe for the negotiations, and also due to the fact that no open consultation took place in advance of the Commission’s proposal. While these two indicators in the process also displayed no or very low levels (for the involvement and visibility of internal and external pro-climate stakeholders), this decreased chances that long-term climate objectives would be considered in the policy process. This links to the final indicator in the policy process – namely the recognition of the functional interrelations between the external gas infrastructure policy and long-term climate policy objectives in the policy discourse. Without this recognition from the policymakers involved (and without them being pushed to make such a connection by pro-climate stakeholders), even the overarching political commitment to climate change at this time was insufficient (and, in this case, it seems it was irrelevant) to advance CPI.

In summary, the very low levels of CPI in the process dimension can help explain why similar very low levels of CPI were found in the output of external gas infrastructure support under the TEN-E guidelines and the 2009 EEPR Regulation. It is the lack of the recognition and articulation of the functional interrelations between this and long-term climate policy that seems most crucial to understanding these very low levels. Additionally, the policy processes in these cases show that policymakers cannot rely on external stakeholders to push them to recognise and act upon the functional interrelations between policy sectors and long-term climate policy objectives.

6.4.5 Overview of explanatory framework While much of the past literature on EPI in the EU highlights the pivotal role of political commitment in advancing policy integration, what is clear in this case study is that political commitment is not a sufficient enabler for CPI (see, e.g., Jordan & Lenschow, 2010). Rather, it seems to be the variable describing the nature of the functional interrelations between external gas infrastructure and long-term climate policy objectives, combined with the way or whether these functional interrelations

267 CPI into EU external gas infrastructure policy are present in the policy discourse that seems to preclude any advances of CPI in the policy process or output.

Table 6.4 below summarises the role of each of the explanatory variables in explaining the level of CPI in both the policy process and the policy output of external gas infrastructure policy in the 2003 and 2006 TEN-E guideline revisions and the 2009 EEPR Regulation. While many of the variables display similar properties across the policy measures, there is a distinct difference between the political commitment to combating climate change generally in the early- to mid-2000s, when the TEN-E revisions were being negotiated, and the late 2000s when the EEPR Regulation was being negotiated. Although overarching political commitment to combating climate change in the late 2000s was high (and higher than for the time of the TEN-E negotiations), this did not push for further CPI. In fact, CPI in 2009 EEPR was very low or even non-existent, despite this overarching political commitment.

Explanatory variables TEN-E (2003, 2006) EEPR (2009) 1. Functional interrelations Indirect & conflictual Indirect & conflictual (process and output) 2. Political commitment (process and output) - To climate change Medium High

- To CPI None or low None or low

3. Institutional and policy context Unfavourable to CPI Unfavourable to CPI (process and output) 4. Process dimension Lack of CPI in process Lack of CPI in process does not promote CPI does not promote CPI (output only) in output in output Table 6.4: Summary of explanatory variables for CPI in external gas infrastructure policy in the 2003 and 2006 TEN-E Decisions and the 2009 EEPR Regulation. In both cases, the functional interrelations are indirect and conflictual – just by the very nature of the policy objectives that promote further gas infrastructure projects. Perversely, it seems that these functional interrelations may be somewhat hidden from policymakers due to the perceived short-term benefit of moving away from coal- generated electricity to gas-powered electricity generation. The long-term conflictual functional interrelations may thus be hidden by short-term considerations, but also by

268 CPI into EU external gas infrastructure policy the surrounding institutional and policy contexts that allow little room for long-term climate concerns to gain traction in external gas infrastructure policy discourses. The history of TEN-E guidelines, for instance, does not promote a dramatic policy change that allows for the abandonment of policies in support of external gas infrastructure (as could be conceived under very high levels of CPI). Neither do procedures (where they existed) for external pro-climate stakeholders to access the policy process guarantee that the functional interrelations will be articulated or acted upon.

It seems, therefore, that the crucial explanatory variables in this case are the nature of the functional interrelations and the (lack of) recognition of these interrelations throughout the policy process. In fact, the nature and lack of recognition of the functional interrelations may be a first-order explanatory variable in this case that cancels out any explanatory value of the remaining variables. The remaining variables – political commitment to combating climate change and to CPI and the institutional and policy contexts – provide more of a background context and further nuance for understanding the levels of CPI in the policy output and process of external gas infrastructure policy in TEN-E and EEPR between 2000 and 2010.

6.5 Conclusions In this chapter, I examined CPI in the policy output and process of policies in support of external gas infrastructure between 2000 and 2010. These policies included explicit financial and political backing provided, at the EU-level, through the 2003 and 2006 revisions of the TEN-E guidelines and the 2009 EEPR Regulation. After introducing the context around external gas infrastructure in the EU, I outlined the benchmarks for very high levels of CPI in external gas infrastructure policy output and process, against which the specific policies under examination could be measured. In the discussion on the external gas infrastructure policy of the TEN-E guidelines and the EEPR Regulation, it was clear that very low levels of CPI (or no CPI) were evident in these cases. To explain these results, I applied the explanatory framework, described in chapter three, to the two policy measures. This meant examining the nature of the functional interrelations between climate policy objectives and external gas infrastructure policy; the political commitment to combating climate change generally and to CPI in particular; and the institutional and policy context in which decisions were made to explain the level of CPI in both the policy output and process. I also

269 CPI into EU external gas infrastructure policy looked for explanatory value within the process dimension to explain CPI in the policy output.

As a result of this analysis, it became clear that the indirect and conflictual nature of the functional interrelations between external gas infrastructure policies and long-term climate objectives blocked any advancement of CPI. This was because the indirect and conflictual nature of the functional interrelations helped to prevent climate considerations entering the policy discourse in the first place. The explanatory power of this variable was buttressed by very low levels of pro-climate stakeholder involvement in the policy process (meaning that any chances of explicit mention of CPI were even further hampered). The remaining explanatory variables provided further background context for understanding the level of CPI in the policy output and process of the EU’s external gas infrastructure policy, but the functional interrelations provided the key explanation in this case.

Such an analysis leads to several conclusions. First, it seems that the nature of the functional interrelations is a first-order explanatory variable for CPI. Where the nature is indirect and conflictual, extra efforts (perhaps through institutional measures) may be required to ensure they are recognised and acted upon by policymakers. Second, it is clear that policymakers cannot rely on external pro-climate stakeholders to push policy in a direction that is more in synergy with 2050 decarbonisation goals. These, and conclusions on all three cases, are discussed in more detail in the next chapter.

270

Conclusions

This concluding chapter provides an overview of the main findings and draws out implications from this research project to contribute to broader discussions of environmental and climate policy integration. The chapter is structured as follows. First, I recall the research aim and the research design. In the second section, I discuss the main findings from the empirical analysis. I present a number of conclusions and implications from the case study analysis. Third, I discuss some of the broader implications of this research project for academic literature on EPI and CPI. Fourth, I describe some new avenues of research that could potentially be explored in follow- up to this project. The findings from this research can be clustered around the case- specific results (including the levels of CPI found, any changes in those levels over time within a particular case and differences or similarities among the cases) and on the explanatory framework (including the relative value of the four explanatory variables applied here).

7.1 Research aim and design This research project was designed to answer the following two-step research question:

What is the extent of climate policy integration in the EU’s energy policy, and why?

Answering this question required, first, measuring the extent of CPI in EU energy policy, and second, applying an explanatory framework with a number of variables to establish why such a level of CPI exists.

In chapter one, I described the background puzzles that inspired this research question, including the role of the energy sector in causing climate change, the adoption of an “integrated” climate and energy package of policy measures at the EU level in 2009 and the EU’s ambitions for leadership on climate change at the international level. These developments aroused curiosity about the reality of CPI in the EU’s energy sector. Some initial guiding questions included: just how integrated are the EU’s

271 Conclusions climate and energy policies? Has the level of CPI changed over time? What can explain potential variation in the level of CPI in EU energy policies? One of the main hypotheses stemming from the research question suggested that varying levels of CPI within the EU’s energy sector could be expected. Thus, I did not expect to find uniform application of CPI across cases of energy policy. A second hypothesis was that the EU’s climate and energy policies (overall) were not well or sufficiently integrated, even if there were examples of energy policy measures where certain levels of CPI did appear.

Little academic work exists on the specific issue of CPI into the EU’s energy policy. There are several studies analysing CPI at national level (e.g., Kivimaa & Mickwitz, 2006; Mickwitz et al., 2009), and much conceptual work on the concept of environmental policy integration (EPI) that provides a natural home for further conceptual development of CPI (Jordan & Lenschow, 2008b; Lenschow, 2002b; Persson, 2007). Discussion of CPI (often considered synonymous to “climate mainstreaming”132) among EU policymakers has been occurring, especially in the context of adapting to the effects of climate change (European Commission, 2009a, 2013a), but academic literature is only beginning to catch up (Adelle & Russel, 2013). By focusing on this particular topic and aiming to answer the above research question, this dissertation contributes to advancing the scholarly discussion about CPI in the EU, while also linking to the policy debates on the issue.

In this project, I examined the level of CPI (measured on a fivefold scale) in the policy process and output of three energy policies in the EU. These three policies were renewable energy policy (chapter four); energy performance of buildings policy (chapter five); and external gas infrastructure policy (chapter six). Each of these three policy areas differs with regard to the objectives, measures and actions in place. Renewable energy policy involves increasing domestic supply of new sustainable energy sources. Energy performance of buildings policy targets the demand side of the energy sector, and aims to reduce energy consumption in the EU. Policies to promote importing natural gas infrastructure deal with external energy relations and aim to ensure stable supplies of imported gas.

132 In the Commission, DG Climate Action policymakers outline that mainstreaming climate policy objectives into other policies is an explicit strategy, see: http://ec.europa.eu/clima/policies/brief/mainstreaming/index_en.htm, accessed: 22 May 2013.

272 Conclusions

The case studies were selected using a most-similar case selection strategy (Gerring, 2008; see chapter one). After an initial survey of the universe of potential cases, these three were chosen as they are drawn as examples from the different types of EU energy policies, and due to the expected variation on the dependent variable (namely, the extent of CPI). The initial survey of the cases resulted in expectations that the RE case would demonstrate high levels of CPI; the EPB case would show medium levels of CPI; and the external gas infrastructure case would result in low levels of CPI (see Table 7.1). Additionally, in each of these cases, policy developments had occurred between 2000 and 2010, allowing the examination of changes over time. Each case study was analysed in-depth using process tracing techniques, with data drawn from document analysis, literature review and interviews (see chapter one; Checkel, 2008; Hopf, 2004; Klotz, 2008).

Drawing on a conceptualisation of CPI that considers it to be a matter of degree (Bryner, 2012), I operationalised the concept by outlining a benchmark for very high levels of CPI. This very high level can correspond to suggestions that environmental or climate policy objectives should be assigned principled priority over the objectives of the policy sector into which they are being integrated (Lafferty & Hovden, 2003). When no weight is assigned to (in this case) climate objectives, Liberatore (1997) argues that there is a risk for “policy dilution”. Hence, the benchmark for very high levels of CPI in the policy output was developed with reference to the long-term objective of reducing GHG emissions in the EU by 80 to 95 per cent by 2050. For each case study, the question could then be asked (with reference to the many studies on the road to decarbonisation by 2050): what role will this policy sector play in achieving decarbonisation by 2050? If the policy output places that sector on such a trajectory, it could then be considered to display evidence of very high levels of CPI. For the policy process, the very high benchmark is rather more qualitative. I assess CPI in the process with reference to three factors, namely the involvement of pro- climate stakeholders internally to the EU policymaking process (such as DG Environment, the ENVI committee in the Parliament, the environment formation of the Council); the involvement of external pro-climate stakeholders in the policymaking process (such as environmental and climate NGOs, pro-climate policy industries and industry associations, including the RE and insulation industries); and the recognition of the functional interrelations between the policy being negotiated

273 Conclusions and long-term climate policy objectives. Each of these three factors was measured on a qualitative fivefold scale to arrive at an aggregated score for the level of CPI in the policy process (see chapter three).

The next step in answering the research question required explaining the results found from the measurement of CPI in the policy process and output in each case. As discussed in chapter three, I used four explanatory variables, namely: functional interrelations; political commitment; institutional and policy context, each of which could help explain the level of CPI in both the policy process and output; and the process dimension, which could help explain the level of CPI in the policy output only. In deriving these explanatory variables, I drew inspiration, first, from EPI literature and its many approaches and (often long) lists of variables (see chapter three), and, second, from general theories of European integration (Fioretos, 2011; Haas, 1958; Moravcsik & Schimmelfennig, 2009; Pollack, 2009; Strøby-Jensen, 2007). Using these perspectives and theories to provide inspiration for the explanatory framework led to the development of a framework that is manageable and applicable to the EU level. I did not set out to test the validity of any of the individual theories, but rather drew from them to understand better the empirical reality of CPI in EU energy policy.

The nature of the functional interrelations between the sector under analysis and long-term climate policy objectives may help in understanding the resulting levels of CPI. Where functional interrelations (meaning how one policy measure or objective could functionally affect the achievement of another policy objective) are direct and synergistic, these interrelations are more likely to enter the policy discourse. Where the functional interrelations are indirect (or not obvious), it may be more difficult for policymakers to recognise them in the policy discourse. Additionally, if the two policies relate in a conflictual manner with each other it may also prove difficult to consider and negotiate on the functional interrelations in the policy process, as trade- offs may be required. Thus, as per Table 3.4, it is expected that where functional interrelations between the policy under analysis and long-term climate policy objectives are direct (or obvious) and synergistic, most favourable conditions exist for the advancement of CPI in both the policy process and output. However, if the functional interrelations are both indirect and conflictual, least favourable conditions exist for advancing CPI.

274 Conclusions

When applying the explanatory variable of political commitment, I analysed two aspects. First, I assessed the overarching political commitment to combating climate change generally in the EU. Second, I assessed the political commitment to advancing CPI in sectoral policies. Political commitment is often considered a crucial variable for pursuing EPI in the academic literature (Jordan & Lenschow, 2010). It is therefore interesting to assess the role this variable plays for understanding CPI. Through the empirical analysis, it may be possible to assess whether or not the role of political commitment is crucial to advancing CPI.

The surrounding environment within which the policy is developed is captured by the third explanatory variable, namely the institutional and policy context. Institutionally, this variable zooms in on the legal basis under which policy was proposed in the EU; whether policy was decided through the co-decision procedure; what sort of decision making procedure (i.e. voting rule) was used; any institutional changes that may have occurred over time; and, whether past policy decisions led to certain policy frameworks being followed. Politically, the wider, external context for policy development may also prove significant for the CPI story. This includes taking account of external variables during the time of policy development, such as: the agenda of international climate negotiations, energy crises, and economic or financial crises. These external developments may have an impact on internal policymaking in the EU, and on CPI.

Finally, the process dimension can help explain CPI in the policy output. For this variable, there is an assumption that there is (usually) a link between the policy process and its negotiations, and the final policy output (Briassoulis, 2005). The level of CPI in the process may thus play an explanatory role for understanding the level of CPI in the policy output. CPI in the policy process is measured in terms of three factors (see above), each of which are measured using indicators over a fivefold scale. Without policymakers first considering and negotiating CPI in the policy process, it would be surprising (though not impossible) to suddenly find evidence of greater levels of CPI in the policy output (ibid.).

7.2 Main research findings In this section, I synthesise, first, the results in answer to the first-part of the research question (“what is the extent of CPI in the EU’s energy policy?”). Second, I focus on

275 Conclusions the explanatory framework and draw conclusions on how to explain the levels of CPI found in the case studies (“and why”?). These results discuss the levels of CPI found in the cases, any changes over time, the differences and similarities among the cases, and the results of the application of the explanatory framework.

7.2.1 Results of the analysis of the levels of CPI In response to the first part of the research question, different levels of CPI in EU energy policy between 2000 and 2010 were found. The assumptions that underlined the initial case study selection were not necessarily confirmed, however. Variation on the dependent variable (the extent of CPI) was expected and confirmed, but not to the same levels as expected (see chapter one and Table 7.1). Where high levels of CPI were expected in the RE case, the levels of CPI in 2001 turned out to be low in the policy output and low to medium in the policy process. In 2009, the level of CPI in the policy process had increased to medium to high but the level of CPI in the policy output had barely improved, measuring low (to medium) (see chapter four and Table 4.2). Where medium levels of CPI were expected in the EPB case, the levels of CPI in the policy process of both the 2002 and 2010 Directives were actually low to medium. The levels of CPI in the policy output of both the 2002 and 2010 Directives were low. There was no significant increase in the level of CPI in the EU’s EPB policy in the first decade of the twenty-first century. Finally, low levels of CPI were expected for the external gas infrastructure policy. In this case, expectations were relatively accurate; although the surprising story is that there is a lack of evidence of any CPI in this case. Levels of CPI in the policy output and process of the external gas infrastructure policy measures of the TEN-E guidelines were none or very low. These levels did not change in the EEPR Regulation in 2009, which also showed none or very low CPI in its policy process and policy output. Table 7.1 summarises the results of the case study analysis of chapters four to six.

276 Conclusions

Initial expected CPI in the policy CPI in the policy Case result process output 2001 RES-E Directive High Low to medium Low 2009 RE Directive High Medium to high Low (to medium) 2002 EPBD Medium Low to medium Low 2010 EPBD recast Medium Low to medium Low 2003, 2006 TEN-E guidelines Low None/very low None/very low 2009 EEPR Regulation Low None/very low None/very low Table 7.1: Summary of levels of CPI in the policy process and output of the case studies, compared to the initial expected results.

Among the more surprising results from this analysis may be the low levels of CPI in the policy output of the RE case, which did not much improve in the 2009 RE Directive, even though levels of CPI in the policy process were medium to high at this time. For each case, however, the results were less than sufficient for achieving 2050 climate policy objectives – even the highest results (in RE) were far from the very high levels of CPI that the 2050 target implies. A clear finding of the analysis is thus that:

• CPI is insufficient in these EU energy policies to achieve long-term climate policy objectives.

A second research finding from these results confirms the underlying hypothesis of the research question that:

• The level of CPI varies among EU energy policy measures.

While this hypothesis is confirmed, the expected levels differed somewhat from the actual levels found (see Table 7.1). This second finding points to a lack of integration even within the energy policy sector in the EU. The three policy objectives of EU energy policy – sustainability, competitiveness and security – are not weighted equally across the various energy policies. Indeed, there seems to be little overlap and discussion among policymakers on the different policy files (interviews 9 & 12), with different policies emphasising one or two of the energy policy objectives. The weight applied to each of the three policy objectives also seems to shift over time, depending on external events, perceived threats and opportunities, and political alliances

277 Conclusions

(interviews 25 & 26). Thus, we can say that even as climate policy objectives are insufficiently integrated into EU energy policy, so is EU energy policy insufficiently integrated internally.

Third, the results show little change in the levels of CPI in the cases over time. This lack of, or little, improvement in the levels of CPI in the three cases studied points to another reality of EU climate and energy policymaking: policymakers are engaged in a game of “catch-up” governance or policymaking, rather than in agreeing sufficiently ambitious policy measures from the outset. In other words, as past policies fail or are shown to be insufficient, the business-as-usual (BAU) scenario for the next policy development shifts. Had the indicative target in the 2001 RES-E Directive been met, for example, more ambitious targets may have feasibly been adopted for the 2009 RE Directive, as the effort to increase the share of RE would already have begun. Instead, the 2009 RE Directive is not ambitious enough to put the EU on the road to decarbonisation within the agreed timeframe (to 2050), as it catches up on the earlier policy failures. The same can be said of the EPB case, where the poor implementation of the 2002 EPBD meant that in the 2010 EPBD recast, policy measures were simply trying to catch-up with the ambitions set eight years previously. For both of these cases, the shifting BAU scenario in the later stages of the first decade of the twenty- first century gave less time for the policy measures agreed to demonstrate effective action on the road to decarbonisation.133 This finding can be summarised as follows:

• Climate and energy policy development in the EU could be characterised as “too little, too late”, when long-term climate policy objectives are concerned.

Fourth, this finding links to an overall indication that the long-term policy perspective to 2050 is not sufficiently considered in policymaking. Had the long-term climate policy objectives been seriously considered in the policy development of the three cases examined, more ambitious policy measures would have been expected. Instead, the insufficiency of CPI in the three cases is striking. The results are, in each of the three cases, lower than what was expected at the outset. It seems that long-term policy perspectives did not play a role in the policy developments to motivate policymakers to agree to more ambitious policies or higher levels of CPI. This finding is discussed

133 In the external gas infrastructure policy case, long-term climate policy objectives never entered the discussions.

278 Conclusions also below, with regard to the recognition of functional interrelations. As a finding of the empirical research, it can be summarised here as follows:

• The lack, or low levels, of consideration of the long-term climate perspective in the three cases is likely to have hindered the advancement of CPI.

Fifth, the differences and changes in the levels of CPI in the policy process of some of the three case studies over time seem to point to a further finding and conclusion. In the RE case, it is particularly striking that the level of CPI in the policy process increased over time, but this did not result in a corresponding increase in the level of CPI in the policy output. For the EPB case, the level of external pro-climate stakeholders involved in the process increased over time, but this also was insufficient to push for higher levels of CPI overall in the process and output (see Table 7.6). In the external gas infrastructure case, CPI in the policy process was very low or non- existent, with external pro-climate stakeholders not involved. These results lead to a consideration about the general “openness” of the EU energy policy processes. In the later years of the 2000s, EU institutions were generally more open to external stakeholders than in the early years of the same decade. Yet, this general openness seems to be insufficient to ensure that climate voices enter the policy discussion. In other words, policymakers may not be able to assume that the interest and involvement of such stakeholders will be sufficient for pushing CPI, or that a general openness to stakeholders will ensure pro-climate stakeholders will be involved. Other ways of advancing CPI in the policy process, and, also in the policy output, may need to be put in place (see also the discussion in the epilogue).

• EU institutional openness to external pro-climate stakeholder involvement in policymaking may be insufficient to advance CPI in the policy process and output.

7.2.2 Findings on the application of the explanatory framework Through the empirical study of chapters four to six, it became clear that certain variables played a greater role in certain cases, meaning that the four variables did not carry equal explanatory weight. Each of the four explanatory variables played more or less crucial roles in the explanation of the levels of CPI found. From this analysis, it may be possible to identify a hierarchy of importance among the explanatory

279 Conclusions variables for an analysis of CPI. In addition, some of the variables proved most valuable as explanatory variables only in combination with others or in support of other variables. In this section, I describe the main conclusions drawn from the application of the explanatory framework in this project. As a reminder, Table 7.2 provides an overview of explanatory variables for each of the cases.

Political Nature of Political Institutional commitment Process Case functional commitment and policy to climate dimension interrelations to CPI context change Low guaranteed Slightly more 2001 RES-E Direct & Medium- involvement Medium favourable to Directive synergistic high of pro- CPI climate stakeholders 2009 RE Direct & Crucial role High High Neutral Directive synergistic of Council More Direct & Crucial role 2002 EPBD Medium Low-medium unfavourable synergistic of Council to CPI More 2010 EPBD Direct & Crucial role High Low-medium unfavourable recast synergistic of Council to CPI Lack of 2003, 2006 Indirect & Unfavourable evidence of TEN-E Medium None or low conflictual to CPI CPI in guidelines process Lack of 2009 EEPR Indirect & Unfavourable evidence of High None or low Regulation conflictual to CPI CPI in process Table 7.2: Summary of explanatory variables for levels of CPI in the case studies. 7.2.1.1 The recognition (and articulation) of functional interrelations as a first-order variable for CPI In each of the case studies, the recognition of the (nature of the) functional interrelations between the policy being developed and the long-term objectives of climate policy proved a crucial first-order variable for advancing CPI. This is most spectacularly represented in the case on external gas infrastructure policy (chapter six). In this case, the nature of the functional interrelations was indirect and conflictual, meaning that least favourable conditions existed for the recognition of these

280 Conclusions interrelations in the policy process (see Table 3.4). The functional interrelations were indirect as the support for external gas infrastructure does not directly lead to increases or decreases in GHG emissions. Several steps are required before the interrelations become clear, including a long-term perspective that such gas infrastructure may lead to the EU becoming locked-in to carbon infrastructure and the recognition that immediate energy security objectives could be achieved through other means. The lack of the long-term perspective (see also above) means that the indirect nature of the functional interrelations between climate policy and external gas infrastructure policy may not have been recognised by policymakers. Secondly, the recognition of the conflictual nature of the functional interrelations again relies on a clear understanding of the long-term risks associated with external gas infrastructure policy. The risk here is that, with such infrastructure operational for fifty years, the energy infrastructure put in place for the future would support the continued use of fossil fuels beyond the time when the EU should be achieving its decarbonisation goals. Thus, policies that support external gas infrastructure conflict with long-term climate policy objectives by emphasising a fossil-fuel infrastructure, rather than promoting a transition to decarbonisation by 2050. These indirect and conflictual functional interrelations reduce chances that they will be recognised by policymakers and dealt with during the policy process. Indirect interrelations may be hidden by their distance and by the steps required to recognise them, and conflictual interrelations may prove difficult to resolve and hence are rather brushed aside by policymakers. Combined, these elements result in little or no demand for CPI in the policy process by policymakers.

In the RE and EPBD cases, the nature of the functional relations were most favourable to advancing CPI, in that they were direct and synergistic. This means that policies to advance the share of RE or to improve the energy performance of buildings can directly help achieve climate policy objectives – they can result in reduced GHG emissions and set the EU on the path towards decarbonisation. For both these cases, the nature of these functional interrelations were recognised and articulated by policymakers and stakeholders in the policy process. The nature of the functional interrelations allowed for easy recognition and discussion in the policy process. This situation led to a demand for CPI (although this was not always explicitly phrased as CPI). However, even in these cases, there was insufficient

281 Conclusions emphasis on the long-term perspective, meaning that even when the functional interrelations are direct and synergistic, an explicit effort to consider the long-term may still be required. From there, the remaining explanatory variables entered the story to explain the final levels of CPI found in the end. Without the recognition and consideration of the functional interrelations, we could hypothesise that it may be less likely to find evidence of CPI in the policy process and output of a certain policy measure.

Recognition of Nature of CPI in the CPI in the policy functional Case functional policy process output interrelations in interrelations process Direct & 2001 RES-E Low to medium Low Medium synergistic Direct & 2009 RE Medium to high Low (to medium) High synergistic Direct & 2002 EPBD Low to medium Low Medium synergistic Direct & 2010 EPBD Low to medium Low Medium synergistic 2003, 2006 Indirect & None/very low None/very low None/very low TEN-E conflictual Indirect & 2009 EEPR None/very low None/very low None/very low conflictual Table 7.3: Extent of CPI in the policy process and output of the cases, the nature and the recognition of these functional interrelations in the policy process (part of the process dimension variable).

Comparison across the three cases therefore shows that where any levels of CPI are found, there is also a certain amount of recognition and articulation/consideration of the functional interrelations between that policy measure and long-term climate policy objectives. Table 7.3 compares the nature of the functional interrelations between each of the cases and long-term climate policy objectives with one element of the process dimension, namely the recognition of these functional interrelations in the policy process. Some correlation with the recognition of functional interrelations and the level of CPI in the policy process and output may be identified in the Table –

282 Conclusions in the case where there was no recognition of the functional interrelations, the level of CPI was found to be none or very low.

7.2.1.2 Political commitment as an insufficient explanation for levels of CPI Much of the literature on EPI highlights the importance of political commitment (or “political will”) for the advancement of EPI (see, for example, EEA, 2005; Jordan & Lenschow, 2010, among others). When it comes to assessing CPI (which can perhaps be considered an example of EPI), it is reasonable to expect that political commitment would also play a crucial role.

For the purposes of this study on CPI in particular, I assessed, first, the political commitment of the EU to combating climate change in a general or overarching sense. This dimension of political commitment could be found in statements, policy documents, and political resolutions or conclusions of the EU institutions, but also in the willingness of the EU to agree internal policy measures to combat climate change (see chapter three). In general, the level of political commitment in this overarching sense was found to have increased between 2000 and 2010, from medium levels in the early-2000s to high levels in the late-2000s. Second, I assessed the level of political commitment to advancing CPI in other policy sectors. This was particular to each of the case studies. In the RE case, political commitment to CPI was medium to high in 2001 and high in 2009. For the EPB case, political commitment to CPI was found to be low to medium in both 2002 and 2010. In the external gas infrastructure case, the political commitment to CPI was none or low in both the TEN-E and the EEPR cases (see Table 7.4).

Taking account of these findings on the political commitment explanatory variable in line with the overall results of the cases (see Tables 7.1 and 7.2) provides a picture that is not quite what would be expected given the emphasis on political commitment in much of the EPI literature. Table 7.4 highlights the role of this variable for each case. It seems that the overarching commitment to combating climate change does not correlate strongly with the levels of CPI found in the three cases. The extreme example is demonstrated by the TEN-E and EEPR cases – while medium or high levels of political commitment to combating climate change existed generally in the EU at the time of these policies’ development, the extent of CPI found in the processes and outputs of these cases was none or very low. Political commitment in

283 Conclusions an overarching sense did not push CPI in these cases. Neither does there seem to be a strong link between overarching commitment to combating climate change and the levels of CPI in the RE and EPBD cases. Even when overarching political commitment to combating climate change was high in the late 2000s, the extent of CPI in the policy output of the 2009 RE Directive was low (to medium); low in the policy output of the 2010 EPBD recast; and low to medium in the policy process of the 2010 EPBD recast. The high levels of overarching commitment to combating climate change may have had some link with the medium to high levels of CPI found in the policy process in the 2009 RE Directive, but such a link is difficult to identify. More likely, this overarching political commitment may have played a reinforcing role for other variables. One potential link could be with the levels of political commitment to advancing CPI in different policies. It may be plausible that higher overarching levels of political commitment to climate change could help push political commitment to CPI in different policy sectors.

Political Political CPI in the policy CPI in the policy Case commitment to commitment process output climate change to CPI Medium to 2001 RES-E Low to medium Low Medium high 2009 RE Medium to high Low (to medium) High High 2002 EPBD Low to medium Low Medium Low-medium 2010 EPBD Low to medium Low High Low-medium 2003, 2006 None/very low None/very low Medium None or low TEN-E 2009 EEPR None/very low None/very low High None or low Table 7.4: Extent of CPI in the policy process and output, and levels of political commitment to combating climate change, generally, and to advancing CPI, in each case.

The level of political commitment to advancing CPI specifically may, thus, be a more important dimension of political commitment as an explanatory variable for the extent of CPI in the EU’s energy policy. This variable displayed different levels in the case study analysis. For external gas infrastructure policy, none or low levels of political commitment to CPI corresponded to none or very low levels of CPI in the policy processes and outputs from 2000 to 2010. It is possible that some correlation exists in

284 Conclusions this case. In the EPBD case, low to medium levels of political commitment to advancing CPI links to low to medium and low levels of CPI in the policy process and output throughout the same decade. It seems in this case that the low to medium levels of political commitment to advancing CPI were insufficient to boost the final results. For RE policy, high levels of political commitment to advancing CPI in 2009 did not seem to increase levels of CPI in the policy output. However, for the same time phase, the levels of CPI in the process were medium to high. It is possible that the high levels of political commitment to advancing CPI played a role in the level of CPI in the process, but that this commitment did not translate into an ambitious policy output. In 2001, while the level of political commitment to advancing CPI in RE policy was medium to high, the overall levels of CPI in the process and output were found to be low to medium and low. It is difficult, therefore, to describe any clear causal link, beyond some hypothesised correlations, between the variable of political commitment to combating climate change and to advancing CPI and the final levels of CPI in the policy process and output of EU energy policy (see Table 7.4).

In sum, the political commitment variable can be considered an important, but insufficient, part of the explanatory story for understanding levels of CPI. In the absence of any political commitment to combating climate change and to advancing CPI, we can imagine that final levels of CPI in the policy processes and outputs of the cases may have been even lower. Nevertheless, political commitment as an explanatory variable plays a role as a second step – once the nature of the functional interrelations is recognised and articulated in the policy process, political commitment may then help explain the final level of CPI found. Additionally, it is a variable that cannot provide complete explanation. Other variables must interact with political commitment for a greater understanding of why such levels of CPI were found. Therefore, it is clear that political commitment by itself is an insufficient explanatory variable (and an insufficient enabler) for CPI.

7.2.1.3 Institutional and policy context variable provides the background context For each of the case studies, I assessed the context within which policies were negotiated and decisions were made. These are time-specific contextual issues that may have affected the level of CPI in EU energy policy developments. In each case, I described the institutional context, referring specifically to internal, institutional factors within the EU. I also discussed the external policy context, meaning external

285 Conclusions factors that may push or block EU policymakers to agree policy on certain issues, within certain timeframes, and to certain levels of ambition.

The main institutional variables (as discussed below) include:

• The legal basis of the proposed policy, the application of the co-decision procedure and the voting rule thus applied (QMV in the Council); • Procedures for consultation, stakeholder involvement and impact assessment; • The role (if any) of EU enlargement (in 2004 and in 2007); • The past policy framework; • The negotiation of related policy instruments alongside the policy being analysed and the role, if applicable, of elections in the Parliament.

All cases were adopted under the co-decision procedure, with various legal bases being applied. The legal bases used included the internal market, the environment and trans-European networks. Qualified majority voting (QMV) in the Council, where no single member state can veto a policy proposal, was the decision-making rule in each case under the co-decision procedure. In the empirical analysis, there was no strong evidence that the QMV rule helped advance CPI. In the case of the 2009 RE Directive, decisions in the Council were explicitly taken under unanimity for political reasons to promote member state buy-in, even though QMV was the official rule that applied (see chapter four). In the EPBD case, the Council was the most unambitious co- deciding institution. It rather worked to water down the ambition of the policy measures. QMV may in theory have helped to prevent the EPBD recast from being watered down further, but it was through negotiations that compromises were made to ensure the recast was finally agreed upon. For the external gas infrastructure policy, QMV was also the decision-making rule for the TEN-E guidelines and the EEPR Regulation, but it does not appear to have helped bring CPI into the policy process or output. In this case, policy change to advance CPI may rather have required the abandonment of the gas aspect of the policy measure – a radical policy change (that may possibly have been easier to achieve under unanimity rules, where just one member state needs to vote against a policy).134 While QMV in theory may help aid

134 Even under unanimity rules, however, a single member state would have to show high political commitment and an understanding of the long-term functional interrelations between climate policy objectives and natural gas infrastructure to make such a radical move in the Council.

286 Conclusions policy change (see chapter two), in the cases examined here, the legal bases and consequent decision-making rule does not seem to have greatly affected the overall level of CPI.

In each case, the openness of the co-deciding institutions to external stakeholders, through consultation procedures and lobbying, evolved over time. As the Parliament gained more power as a co-legislator, it also increased interactions with external stakeholders (many of whom classified access to the Parliament as “easy”; interviews, 5, 6, 7, 8, 11 & 20). The Commission outlined standards for public consultation in 2002 that included obligations for consultation and impact assessment for each policy proposal (European Commission, 2002d). The Commission and Parliament later worked together to develop the European transparency initiative, to enhance openness and transparency in formal and informal lobbying and consultation procedures (European Commission, 2006c, 2007d). For both of these institutions, consultation procedures, impact assessment procedures and general openness to stakeholders improved or increased between 2000 and 2010. The exception is rather with the Council. External stakeholders still described difficulties in accessing the Council (interviews 5, 6 & 7). The Council does not usually need to access external expertise and does not have procedures that oblige it to engage in EU-level consultation procedures. Thus, when it comes to the institutional factor on consultation procedures, there are differences among the three co-deciding institutions and over time. Additionally, although the 2009 EEPR Regulation was agreed during the time when impact assessment and consultation procedures were standard, neither an impact assessment nor a public consultation took place (ostensibly due to the urgent response to the economic crisis required at this time). This institutional factor provides some nuance to, and is part of the background context for, understanding the levels of CPI found in the case studies.

A further institutional context involves the enlargement of the EU from 15 member states to 27. In each of the cases, the enlargement meant that the early policy measures being assessed were negotiated and agreed among the EU15 (the 2001 RES- E Directive, the 2002 EPBD, and the 2003 TEN-E Decision), but later developments were negotiated and agreed among 25 (2004-2006) or 27 (from 2007) member states (the 2006 TEN-E Decision, the 2009 RE Directive, the 2010 EPBD recast and the 2009 EEPR Regulation). This institutional development may be considered to have

287 Conclusions affected in some way the levels of CPI, and the evolution of the levels of CPI in the cases. Nevertheless, it is difficult to pinpoint specific effects related to enlargement in the cases. In the 2009 RE Directive, Poland (a new member state) seemed to be a reticent party in agreeing ambitious policy measures, but achieved concessions in the parallel negotiations on the ETS Directive (2009/29/EC) and did not block the developments on the RE Directive. In fact, Italy was considered a member state that caused more difficulties in the negotiations in this case than any of the newer member states (see chapter four). For the 2010 EPBD recast, the energy efficiency standards in the newer member states were generally lower than those of many of the EU15. However, these new member states were not isolated in their opposition to strengthening the EPBD – among the EU15 there were many member states that were also reluctant to agree (or at least uninterested in agreeing) an ambitious output on this policy file. For the external gas infrastructure policy, the accession of twelve new member states between 2004 and 2007 that were largely dependent on Russia for supplies of gas is likely to have enhanced the importance of energy security objectives in the EU. These member states were thus supportive of measures for new external gas infrastructure, for security of supply reasons. Yet, the other member states, and the Parliament and Commission were also supportive of these policies (to the detriment of CPI). Thus, EU enlargement may add some elements to the institutional background within which EU energy policy is made, but there is no clear evidence in these cases that enlargement played a key role in blocking or advancing CPI. It is rather a background context that can be more or less favourable for the advancement of CPI, but an insufficient explanatory variable.

In most cases, there was some policy framework or past policy measure that led to a continuation of long-held ideas on the policy direction, objectives and tools. For the RE case, there was some policy change from indicative early targets to mandatory targets in the 2009 RE Directive. However, the general policy framework on RE was outlined already in the Commission’s 1997 white paper (European Commission, 1997b). In the EPBD case, the policy framework stems from Commission policy documents of the mid-1980s (European Commission, 1984), when neither the environment nor climate change were part of energy efficiency policy. The same policy framing continued for buildings right through to the 2010 EPBD recast. In the external gas infrastructure policy, the relatively long history of trans-European

288 Conclusions guidelines for energy (TEN-E) in the EU also outlined a certain frame for policy development. In this case, even the suggested gas infrastructure projects remained a part of the policy frame from the late 1990s right through to the 2009 EEPR, seemingly without deeper analysis in light of climate change concerns (see chapter six).

Both the EPBD and RE Directive were negotiated alongside other policy measures. The RE Directive was negotiated as part of the overall climate and energy package, and it seems that a limited amount of negotiating on the policy measures in the package took place (see, e.g. the case of Poland and the ETS Directive). This is very limited, however, and does not seem to have had a clear effect on the level of CPI in the RE Directive. The EPBD recast was first proposed in 2008 when the climate and energy package was being negotiated. Again, it is difficult to pinpoint any direct effect this had on the level of CPI in the EPBD recast. It may perhaps have facilitated climate policy entering discussions at this time, but there is little evidence. The 2009 EEPR was also negotiated alongside negotiations on the EPBD recast and in light of the agreement on the climate and energy package in 2008. There is no evidence that these other policy negotiations linked to discussions on the EEPR Regulation. Additionally, the EEPR Regulation was proposed, negotiated and agreed within the first half of 2009, when many MEPs were preparing their election campaigns. This may have led to the Parliament paying little attention to this policy file (interviews 25 & 26). However, even if the Parliament had been paying close attention to this policy measure, it cannot be said for certain that higher levels of CPI would then have materialised. Generally, in the three cases, the past policy frame within the EU was followed without much room for radical policy change that could advance CPI. The negotiations of other policy measures alongside the policy measure under examination and the election schedule of Parliament do not seem to have had a significant influence on the levels of CPI.

The external policy context also involved several variables that may have affected the advancement of CPI in the cases. The potentially most influential variables included:

• The (schedule of) the international climate negotiations, linked to EU ambitions to demonstrate leadership on climate change;

289 Conclusions

• Growing EU dependence on imports of fossil fuels, and gas supply security issues between 2000 and 2010; • And the financial and economic crises that began in autumn 2008.135

In the RE and EPBD cases, the international climate negotiation schedule was found to increase the pace at which internal policy was made (see also Boasson & Wettestad, 2013). For the 2009 RE Directive, agreement was reached in the first reading in just over ten months from the date of the publication of the proposal. This meant that the agreement was announced during the international climate negotiations in Poznan in December 2008, and in advance of the Copenhagen climate conference scheduled for December 2009. The RE Directive was agreed as part of the climate and energy package that was explicitly meant to add credibility to the EU’s leadership claims for leadership on climate change, by demonstrating ambitious internal policy measures. For the 2010 EPBD, the Swedish Presidency worked to ensure that political agreement was reached just in time for the 2009 climate negotiations in Copenhagen (see chapter five). While it does not seem that the international negotiations agenda boosted the level of ambition of the policy measures, it seems that the agenda added an implicit pressure for the EU to agree on policy measures within a certain timeframe. In the external gas infrastructure case, the international climate negotiations did not feature. Linked to the fact that the functional interrelations between external gas infrastructure support policy and long-term climate policy objectives were not considered in the policy process, climate policy objectives (and the international climate negotiations) were not at all discussed or considered in this case. Thus, this particular factor provides a background context that can explain some nuance in the levels of CPI found, but is not a sufficient explanatory variable by itself.

The energy security issues surrounding the rising dependence of the EU on imports of fossil fuels motivated the earlier TEN-E Decisions, but this security reality was cited as a motivation for policy development for EPB and RE policies also. The 2006 and 2009 gas supply interruptions were part of the motivation for the gas infrastructure policy of the EEPR Regulation. These gas crises did little to motivate policy developments in the RE and EPBD cases, except as examples of the importance of

135 The nature of the buildings sector (with different types of buildings across the EU, and many different local-level actors) was also found as a contextual factor for the EPBD case.

290 Conclusions these measures for achieving energy security also. As an external policy context, the dependence of the EU on imports of energy, and the gas crises were motivating factors that were important for the continuation of the TEN-E guidelines and the development of the EEPR Regulation in support of external gas infrastructure policy.

The financial and economic crises that began in 2008 may also have had some impact on the level of ambition of the policy measures agreed since 2008. The 2009 RE Directive was agreed in December 2008, as the extent of the economic and financial crises was not yet fully known or realised in policymaking circles. The crises, thus, may not be pinpointed as reasons for lower levels of CPI in this case. In the 2010 EPBD recast, the crises amplified member state reticence to agree on costly and ambitious measures, but this is certainly not the only reason for member state hesitation in this case. Combined with the local-level nature of the EU’s building sector and calls for subsidiarity, the potential costs of measures under the 2010 EPBD recast were just one of the issues. For the 2009 EEPR Regulation, the financial and economic crises were the policy’s raison d’être. The Regulation was proposed with the precise motivation of ensuring that investment in energy projects would continue despite the crises. It was pushed through the co-decision procedure very rapidly, with no impact assessment or consultation procedure, as it was deemed an urgent response to the financial and economic crises. For the 2009 EEPR, at least, these crises motivated policy development (with little consideration of CPI).

CPI in the policy CPI in the policy Case Institutional and policy context process output Slightly more favourable to 2001 RES-E Low to medium Low CPI 2009 RE Medium to high Low (to medium) Neutral 2002 EPBD Low to medium Low More unfavourable to CPI 2010 EPBD Low to medium Low More unfavourable to CPI 2003, 2006 None/very low None/very low Unfavourable to CPI TEN-E 2009 EEPR None/very low None/very low Unfavourable to CPI Table 7.5: Extent of CPI in the policy process and output, and role of the institutional and policy context in explaining the levels of CPI.

291 Conclusions

From all these institutional and policy contextual factors, no single issue stands out as providing an overall explanation for the levels of CPI in the cases. Some factors are more relevant in one case than in another. In combination, these factors provide a background context that can be more or less favourable to the advancement of CPI (see Table 7.5). Except in the case of the external gas infrastructure policy, where the contextual factors were found, in combination, to be considerably more unfavourable to advancing CPI, there was no outstanding contextual factor that explained the levels of CPI in RE or EPBD policy. These institutional and policy considerations may rather be understood as providing the background context – they describe the reality of the time in which the policy developments came about, and in some cases may add nuance to the explanatory value of the remaining variables. However, no contextual factor by itself can be considered as a main reason for the levels of CPI found in the policy processes or outputs of the three case studies examined here.

7.2.1.4 Breaking down the process dimension to explain CPI in the policy output With regard to the role of the process dimension as an explanatory variable to understand the levels of CPI found in the policy output, a link was expected between the levels of CPI in the process and output (see above and chapter three). The assumption here is that the policy output usually flows and follows from the negotiations in the policy process. While it is conceivable that CPI could suddenly enter the final agreement on a policy without having been part of the policy discussions, this could be considered unusual (Briassoulis, 2005a). The results (see Table 7.1) seem to validate this expectation in one respect: in no case were the levels of CPI in the output greater than the levels of CPI found in the policy process. This result by itself does not say much about the role of the process in explaining CPI in the policy output, however. To understand better the explanatory role of the policy process, it is necessary to delve into the indicators used to measure the extent of CPI in the policy process. Thus, to see what precisely in the policy process can explain CPI in the policy output, the level of CPI in the policy process needs to be broken down.

292 Conclusions

Involvement Involvement CPI in the Recognition CPI in the of internal of external Case policy of functional policy pro-climate pro-climate output interrelations process stakeholders stakeholders 2001 Low- Low to Low Low Medium RES-E medium medium Low (to Medium- Medium- Medium to 2009 RE Medium medium) high high high 2002 Low to Low Low Low Medium EPBD medium 2010 Low to Low Low Medium Medium EPBD medium 2003, None/very None/very None/very 2006 Very low Very low low low low TEN-E 2009 None/very None/very None/very None/very None/very EEPR low low low low low Table 7.6: CPI in the policy output, compared to the three indicators and the overall level of CPI in the policy process.

Table 7.6 shows the overall levels of CPI in the policy output for each case, along with the levels found for each of the three indicators, which combined provided the overall level of CPI in the policy process. Even at this level of analysis, only a limited understanding of the explanatory role of the policy process comes through. It can be said that the level of CPI in the policy process is linked to the involvement of pro- climate stakeholders and the recognition of the functional interrelations of the policy sector with climate policy objectives. To get more from this explanatory variable, these indicators need to be broken down further, with regard to the policy process in each of the co-deciding institutions of the EU, in particular. In this way, we should be able to discover whether there were certain periods during the policy process, or certain actors, that help explain the levels of CPI eventually found in the policy output.

Except in the external gas infrastructure policy case, where no or very little evidence of CPI in the policy process across all three indicators was found, the role of the

293 Conclusions

Council is shown to be crucial for understanding the levels of CPI, when each indicator is examined more closely. In the RE case, the involvement of internal and external pro-climate stakeholders measured either very low or low across the two time phases. This compares to high involvement of internal and external pro-climate stakeholders in the Parliament in the 2009 RE Directive, for example. Similar very low or low levels of involvement from internal and external pro-climate stakeholders in the Council were found for the 2002 and 2010 EPBDs. In the 2010 EPBD recast, external pro-climate stakeholder involvement with the Parliament measured high. Even when the recognition of functional interrelations in each of these cases measured medium, the Council proved less open to climate voices being raised. Thus, although the policies were agreed under the co-decision procedure, it seems that the Council nevertheless played a more crucial role in the policy process (usually to lower ambition and to prevent very high levels of CPI materialising; see Table 7.2).

This discovery about the explanatory role of the policy process also leads to a consideration on the future development of the conceptual framework. It may be a better reflection of the reality of EU policymaking if the indicators to measure CPI in the policy process were separated among the institutions. Another strategy to adjust the framework could be to assign different weight to the institutions – with the Council’s score on each indicator weighted more than the score for the Parliament and the Commission. This may prove problematic methodologically, and a clear justification (perhaps including empirical evidence) for the weighting system may be required. Yet this strategy may help alleviate the issue that the policy process must be broken down into its component parts before its explanatory value is more fully revealed.

Thus, the conclusion on this explanatory variable is that the level of CPI in the policy process must be broken down for a nuanced explanation of the levels of CPI in the policy output. In the empirical research carried out here, it was the results of the indicators in relation to the Council that played a crucial explanatory role. The process dimension is, again, an insufficient explanatory variable for understanding CPI overall. It may be deployed in concert with the three explanatory variables listed above – which each help explain the level of CPI in the policy output and the policy process. However, the indicator on the recognition of the functional interrelations forms part of the first-order explanatory variable discussed above.

294 Conclusions

7.2.1.5 Hierarchy of explanatory variables Further empirical research may be required to test the explanatory framework and develop a larger evidence base from which to draw conclusions about the hierarchy or constellations of explanatory variables that prove most crucial for a study on CPI. From the analysis of the cases in this study, some initial conclusions can nevertheless be drawn.

First, the nature and recognition of the functional interrelations is a crucial first-order explanatory variable for understanding CPI. This combines the first explanatory variable on the nature of the functional interrelations with one element of the process dimension – namely the recognition (and articulation) of these functional interrelations in the policy process. Second, political commitment plays a role in the explanation of the levels of CPI only once the functional interrelations have been recognised and articulated. Unless there is some recognition of the functional interrelations by the policymakers in a policy sector, political commitment to combating climate change or to advancing CPI is unlikely to play a role in advancing CPI. Third, the institutional and policy context provides a background context that can provide further explanatory nuance, when the nature and recognition of the functional interrelations and the level of political commitment do not seem to provide sufficient explanation. Fourth, the process dimension is most useful as a complementary explanatory variable when broken down into its component elements. One element of the process dimension – namely the recognition of the functional interrelations – is nevertheless part of the first step in the explanation of levels of CPI. Therefore, rather than a hierarchy of variables, we could perhaps describe the variables as interrelating with each other in a loosely sequential manner, and as reinforcing the overall explanation.

7.3 Summary of findings To summarise, the main conclusions that may be drawn from this research project are as follows:

• The level of CPI in the three cases of EU energy policy analysed was found to be insufficient to achieve long-term climate policy objectives to 2050; • The level of CPI varied among the policy measures;

295 Conclusions

• The level of CPI evolved only little or not at all in the cases between 2000 and 2010, pointing to a situation of what could be considered “catch-up governance”; • The lack, or low levels, of consideration of the long-term climate perspective in the three cases may have hindered the advancement of CPI; • The recognition and consideration of the (nature of the) functional interrelations between the policy being developed and long-term climate policy objectives in the policy process seemed to be a crucial first-order variable for the advancement of CPI; • Political commitment, both to combating climate change and to advancing CPI, seemed an important, but insufficient, variable for understanding and/or advancing CPI. Overarching political commitment to combating climate change was not found to push CPI when the functional interrelations were not recognised; • The institutional and policy contexts within which policy is developed seem to have provided a background context that may add nuance to an overall explanation (by providing more or less favourable surrounding conditions for the advancement of CPI) – this was not, by itself, a crucial explanatory variable; • The process dimension was found to play a role in understanding the level of CPI in the policy output, yet this explanatory variable needed to be further broken down for a clearer understanding: the crucial role of the Council was thus revealed for the RE and EPBD cases.

What should be clearly highlighted from these conclusions is a further overarching conclusion:

• No explanatory variable was sufficient on its own to explain the levels of CPI in the policy process and output of EU energy policy.

7.4 Contribution to literature The focus of this study was not on theory testing, but rather on a qualitative empirical analysis questioning the accuracy of describing EU climate and energy policy as “integrated”. The research, nonetheless, contributes to academic discussions on EPI and CPI.

296 Conclusions

The literature on EPI in the EU has quite a long history, but there is continued disagreement on what EPI is and how it can be identified or advanced (Jordan & Lenschow, 2008b; Lenschow, 2002b; Nilsson & Persson, 2003; Nollkamper, 2002; Persson, 2004). CPI, in particular, is still a new topic in academic literature (Adelle & Russel, 2013; Ahmed, 2009; Rietig, 2012), and there is little focus on CPI at the EU level. The theoretical or conceptual contribution of this work to the literature is in distilling the conceptual frameworks developed in the extensive literature on EPI and CPI and, combined with inspiration from general theories of European integration, in deriving a manageable conceptual framework for analysis of CPI in the EU. I developed a conceptualisation of CPI that captures differing levels of CPI in the policy process and output. Conceptualising CPI as a matter of degree allowed for an empirical investigation of the extent of CPI that could, methodologically, capture very high levels of CPI (“strong” CPI) and also very low levels of CPI (“weak” CPI). I did not consider CPI as something that either exists or does not exist. A more nuanced conceptualisation and operationalisation allowed for the reality of different levels of CPI in EU energy policy to be revealed.

Second, this research project proposed a framework for analysing CPI in the EU that is manageable, yet also aims for comprehensiveness. The analytical framework combines variables derived from the long-established EPI literature and several established theoretical perspectives on European integration. Although the explanatory strength of the variables differed depending on the empirical realities of the cases, it can certainly be said that the combination of these variables may have helped explain the empirical reality more fully than a reliance on an explanatory framework derived or inspired from a single theory or perspective. This framework can be tested in different policy sectors, and through its application, can be refined and improved, especially as regards the role of the process as an explanatory variable (see above).

Third, the empirical analysis in this research project included an in-depth assessment of CPI in EU energy policy, and challenges the assumption that the 2009 climate and energy package of measures adopted in the EU represents evidence for the advancement of EPI or CPI. While, on paper, this seems a plausible assumption, it is clear from the empirical evidence presented here that such assumptions are unfounded. Even the EU’s RE policy, with the 2009 RE Directive adopted as part of the climate

297 Conclusions and energy package, displays at best low (to medium) levels of CPI in the policy output from 2000 to 2010. The empirical findings in this project should lead scholars of EPI/CPI to look twice before accepting claims of advanced policy integration in the EU.

7.5 Future research While I set out to answer a seemingly simple question (“what is the extent of CPI in the EU’s energy policy, and why?”), many other questions can be raised as a result of the analysis. Future research could focus on several empirical and conceptual questions.

The conceptual framework developed in this project could benefit from wider application. Through testing and refining the framework, its strengths and weaknesses could be more clearly identified. One weakness of the framework identified through its application in this project is that the explanatory role of the process requires the process dimension to be broken down to levels lower than the overall indicators before a valuable explanatory role may be revealed. Future research could help especially to refine this aspect of the framework. Such future research could take the shape of similar studies in different sectors of EU policymaking (such as transport, industry or agriculture policies). Additionally, the framework could benefit from slight adaptions and testing on other levels of policymaking – does the explanatory framework hold value also for analyses of CPI on the national level or international level? Such empirical research would enhance the evidence base for further theorising of CPI in general.

In practice, CPI seems to be far more appealing to policymakers (also known as “climate mainstreaming” in policy circles) than EPI ever was (Adelle & Russel, 2013). Although this project demonstrated that CPI in the EU’s energy sector seems far from sufficient to achieve long-term climate policy objectives, there is nonetheless a certain level of CPI in some energy policies. The same may not be so easily identified for EPI. EPI may represent a higher level of integration that involves systemic and holistic thinking; it requires that the “environment”, broadly defined, be considered in all policymaking. By focusing on CPI, do academics and policymakers harm the overarching EPI notion? Will this focus lead to further erosion of holistic policymaking? Is climate change too politically salient an issue to allow other

298 Conclusions environmental considerations to enter the discussion? What consequences could a focus on CPI, rather than on EPI, actually mean for the environment?

Many of these questions require philosophical theorising and cannot be answered solely with strong empirical analysis. Future research on CPI may thus require further thinking on these fundamental questions, which are unavoidably normative in nature. In some respects, these questions can be summed up in one: is CPI good for the environment? Some initial theorising on the links between EPI and CPI in the literature has begun, with Adelle and Russel (2013) discussing this specific issue. Taking this theorisation further and questioning the very justifications for CPI, combined with evidence-based empirical studies, may provide useful insights into how priorities should or could be assigned to different objectives in policymaking.

Finally, the issue of the challenges of long-term policymaking is one that recurred throughout this study. Climate change is a complex issue – GHGs emitted today cause (uncertain) effects in the future. With the IPCC’s periodic assessment reports, the scientific consensus states that GHG emissions will have to peak as soon as possible before declining dramatically, on a global scale, in order to avoid potentially catastrophic consequences caused by climate change (IPCC, 2007). Such a global, complex problem, with decisions needed today to avoid future consequences, poses particular challenges for politics and policymakers working within short electoral cycles. This “wicked” problem, as some scholars name climate change (Adelle & Russel, 2013; Jordan, Huitema, van Asselt, Rayner, & Berkhout, 2010), requires a level of political courage and innovation that previously in history was mobilised in situations of war.

For climate policy development in the EU, the challenges of policymaking with objectives on a long-term horizon to 2050 are also present. Several interviewees for this research project mentioned that the 2050 objectives to reduce GHG emissions by 80 to 95 per cent are not considered in day-to-day policymaking. For politicians, the commitment is valid, but the measures to be put in place to reach the goal are not priorities (interviews 1, 4, 5, 6, 10, 23 & 26). In the case studies in this research, a lack of consideration of the long-term perspective in policymaking was evident. Academic discussions on the challenges of solving climate change in democratic systems (Barker, 2008; Jordan, Huitema, & van Asselt, 2010) could be complemented

299 Conclusions with further research on how the EU deals with the climate change issue, having committed itself to the long-term 2050 goal. Describing and analysing EU developments towards achieving long-term goals can be carried out using the 2050 goal as a benchmark, and/or by assessing CPI as carried out here. Linking this work to theories of democratic politics may prove fruitful for broadening the debate and discussion beyond the energy sector and even beyond the EU.

300

Epilogue

Remarks and recommendations for policymakers

This research project analysed EU energy policy from 2000 to 2010. With climate policy integration (CPI), or “climate mainstreaming”, becoming a strategy in the EU to combat climate change, I close this dissertation with some remarks for policymakers interested in advancing CPI. Policymakers should consider the ramifications of the main result: that CPI is insufficient in the three EU energy policies examined here to achieve long-term (2050) climate policy objectives.

First, long-term planning for policy (i.e. the 2050 horizon) has yet to be fully aligned with day-to-day politics and policymaking in the EU. This is a challenge especially for elected politicians working within an electoral cycle. Civil servants (often with the ability to remain longer in policymaking) may need to bear at least some of the burden of pushing to achieve these long-term objectives in practice. A consideration of long-term climate and decarbonisation policy objectives should be made an explicit part of the evaluation of any policy proposal. Where a policy sector objective is not in line with long-term objectives, policymakers may need to consider radical policy options (such as abandoning policies or making innovative adjustments to established policy frameworks). The Commission, in particular, can play the role of advancing discussions on long-term policymaking, for example, through green and white papers and by demonstrating long-term policymaking by example across its proposals.

Second, especially when the nature of the functional interrelations between a policy sector and long-term climate policy objectives are indirect and/or conflictual, explicit efforts will be required to ensure the interrelations are recognised and considered in policymaking. These interrelations must first be recognised before policymakers can consider how to deal with them. A general openness to pro-climate stakeholders seems insufficient to ensure climate policy objectives are considered across policy sectors. This may require procedural rules, such as an explicit requirement to consider and report on the potential functional interrelations with long-term climate policy objectives for every policy measure under development. The requirement in Article 11, TFEU, to integrate environmental protection requirements into the definition and

301 Epilogue implementation of the Union’s policies and activities is a legal step towards integration that has not been translated sufficiently into everyday policymaking.

Third, to meet the decarbonisation goals by 2050, trade-offs may be needed to prioritise the move to decarbonisation, even within energy policies. The three objectives of energy policy – security, competitiveness and sustainability – are not weighted evenly in policymaking and are often not even integrated with each other. To achieve decarbonisation, policies motivated by concerns for energy security, for example, may need to be adjusted in light of decarbonisation objectives. Continuing secure supplies of fossil fuels into the future runs counter to decarbonisation objectives. Security objectives may need to be reframed in terms of decarbonisation- friendly policies (such as enhancing domestic supplies of renewable energy). Furthermore, policies that aim to ensure low energy costs may need to be reassessed. If decarbonisation objectives require upfront capital investment, priority may need to be given to achieving this objective over the objective of ensuring costs are kept to a minimum. This particular trade-off is likely to prove controversial in times of economic recession. Innovative and creative linking of decarbonisation goals to objectives of job creation may be required. Additionally, assigning priority in policymaking to decarbonisation policies may need a high-level political decision, with more explicit commitment to this goal. A coalition of policymakers and politicians may be required to push for such an explicit commitment and decision.

Finally, policymakers’ buy-in to ensuring the integration of climate policy objectives within energy policy decisions is not yet obvious. Interaction among policymakers seems too limited, both formally and informally, to advance CPI (even among energy policymakers focused on different dossiers). Some procedures may need to be adopted to ensure that long-term climate policy objectives are considered across sectors, including obligations for closer collaboration among policymakers on the development of policy proposals. Requirements for long-term climate objectives to be considered in the impact assessment procedure and requirements for obligatory consultation with (internal and external) pro-climate stakeholders, regardless of the topic of the policy being developed may be needed. Such procedures may need to be developed to ensure climate change does not become further compartmentalised within certain committees or DGs (such as within the DG for Climate Action in the Commission).

302

Bibliography

Adelle, C., Pallemaerts, M., & Chiavari, J. (2009). Climate change and energy security in Europe: policy integration and its limits. Stockholm: Swedish Institute for European Policy Studies. Adelle, C., & Russel, D. (2013). Climate policy integration: a case of déjà vu? Environmental Policy and Governance, 23(1), 1-12. Adelle, C., Russel, D., & Pallemaerts, M. (2012). A “coordinated” European energy policy? The integration of EU energy and climate change policies. In F. Morata & I. Solorio Sandoval (Eds.), European energy policy: an environmental approach (pp. 25-47). Cheltenham: Edward Elgar. Ahmed, I. H. (2009). Climate policy integration: towards operationalization. UN/DESA. Alvarez, R. A., Pacala, S. W., Winebrake, J. J., Chameides, W. L., & Hamburg, S. P. (2012). Greater focus needed on methane leakage from natural gas infrastructure. Proceedings of the National Academy of Sciences of the United States of America (PNAS), 109(17), 6435-6440. Andaloro, A. P. F., Salomone, R., Ioppolo, G., & Andaloro, L. (2010). Energy certification of buildings: a comparative analysis of progress towards implementation in European countries. Energy Policy, 38(10), 5840-5866. Bahgat, G. (2006). Europe’s energy security: challenges and opportunities. International Affairs, 82(5), 961-975. Baran, Z. (2007). EU energy security: time to end Russian leverage. The Washington Quarterly, 30(4), 131-144. Barker, T. (2008). Climate policy: issues and opportunities. In H. Compston & I. Bailey (Eds.), Turning down the heat: the politics of climate policy in affluent democracies (pp. 15-32). Houndmills: Palgrave MacMillan. Barnes, P. M. (2011). The role of the Commission of the European Union: creating external coherence from internal diversity. In R. K. W. Wurzel & J. Connelly (Eds.), The European Union as a leader in international climate change politics (pp. 41-57). Abingdon, Oxon: Routledge. Beck, U. (2009). Understanding the real Europe: a cosmopolitan vision. In C. Rumford (Ed.), The SAGE handbook of European Studies (pp. 602-619). London: Sage Publications Ltd. Behrens, A., & Egenhofer, C. (2011). Rethinking European climate change policy. In V. L. Birchfield & J. S. Duffield (Eds.), Towards a common European Union energy policy: problems, progress, and prospects (pp. 217-234). New York: Palgrave MacMillan.

303 Bibliography

Bennett, A. (2008). Process tracing: a Bayesian perspective. In J. M. Box- Steffensmeier, H. E. Brady & D. Collier (Eds.), The Oxford handbook of political methodology (pp. 702-721). Oxford: Oxford University Press. Biermann, F., Davies, O., & van der Grip, N. (2009). Environmental policy integration and the architecture of global environmental governance. International Environmental Agreements: Politics, Law and Economics, 9(4), 351-369. Birchfield, V. L., & Duffield, J. S. (Eds.). (2011). Towards a common European Union energy policy: problems, progress, and prospects. New York: Palgrave MacMillan. Blair, T. (2005). Opening statement: press conference at EU informal summit Hampton Court. Retrieved 6 February 2013, from http://webarchive.nationalarchives.gov.uk/+/http://www.number10.gov.uk/Pag e8393. Boasson, E. L., & Wettestad, J. (2013). EU climate policy: industry, policy innovation and external environment. Farnham: Ashgate. Bowyer, C. (2010). Anticipated indirect land use change associated with expanded use of biofuels and bioliquids in the EU – an analysis of the National Renewable Energy Plans. Brussels: Institute for European Environmental Policy. BP. (2002). BP statistical review of world energy June 2002. London: British Petroleum. BP. (2006). BP statistical review of world energy June 2006. London: British Petroleum. BP. (2011). BP statistical review of world energy June 2011. London: British Petroleum. BP. (2012). BP statistical review of world energy June 2012. London: British Petroleum. BPIE. (2011). Europe’s buildings under the microscope: a country-by-country review of the energy performance of buildings. Brussels: Buildings Performance Institute Europe. Bretherton, C., & Vogler, J. (2006). The European Union as a global actor. London: Routledge. Briassoulis, H. (2005a). Analysis of policy integration: conceptual and methodological considerations. In H. Briassoulis (Ed.), Policy integration for complex environmental problems: the example of Mediterranean desertification (pp. 50-80). Aldershot: Ashgate Publishing Limited. Briassoulis, H. (2005b). Complex environmental problems and the quest for policy integration. In H. Briassoulis (Ed.), Policy integration for complex

304 Bibliography

environmental problems: the example of Mediterranean desertification (pp. 1- 49). Aldershot: Ashgate Publishing Limited. Broscheid, A., & Coen, D. (2007). Lobbying activity and fora creation in the EU: empirically exploring the nature of the policy good. Journal of European Public Policy, 14(3), 346-365. Bryner, G. C. (2012). Integrating climate, energy and air pollution policies (with Robert J. Duffy). Cambridge, MA: MIT Press. Buchan, D. (2009). Energy and climate change: Europe at the crossroads. Oxford: Oxford University Press. Buchan, D. (2010). Energy policy: sharp challenges and rising ambitions. In H. Wallace, M. A. Pollack & A. R. Young (Eds.), Policy-making in the European Union (Sixth ed., pp. 357-379). Oxford: Oxford University Press. Burns, C., & Carter, N. (2010). Is co-decision good for the environment? An analysis of the European Parliament’s green credentials. Political Studies, 58(1), 123- 142. Burns, C., & Carter, N. (2011). The European Parliament and climate change. From symbolism to heroism and back again. In R. K. W. Wurzel & J. Connelly (Eds.), The European Union as a leader in international climate change politics (pp. 58-73). Abingdon, Oxon: Routledge. CANEurope. (2011). Energy efficiency and savings: clearing the fog. Everything you always wanted to know but were afraid to ask about Europe’s first energy source. Brussels: Climate Action Network – Europe. Carter, N., & Burns, C. (2009). Is the European Parliament an environmental champion? Full research report, ESRC end of award report, RES-000-22- 2304. Swindon: Economic and Social Research Council. Checkel, J. T. (2008). Process tracing. In A. Klotz & D. Prakesh (Eds.), Qualitative methods in international relations: a pluralist guide (pp. 114-127). New York: Palgrave Macmillan. Cini, M. (2007). Intergovernmentalism. In M. Cini (Ed.), European Union politics (Second ed., pp. 99-116). Oxford: Oxford University Press. Corbett, R., Jacobs, F., & Shackleton, M. (2007). The European Parliament. London: John Harper Publishing. Corbett, R., Peterson, J., & Bomberg, E. (2012). The EU’s institutions. In E. Bomberg, J. Peterson & R. Corbett (Eds.), The European Union: how does it work? (Third ed., pp. 47-73). Oxford: Oxford University Press. Council of the European Union. (1986). Resolution concerning new Community energy policy objectives for 1995 and convergence of the policies of the Member States. No: C 241/1 (16 September 1986).

305 Bibliography

Council of the European Union. (1996). 1939th Council meeting: environment. No: 8518/96 Presse: 188 (26 June 1996). Council of the European Union. (1997). 2009th Council meeting: energy. No: 8319/97 Presse: 167 (27 May 1997). Council of the European Union. (2001). Common position adopted by the Council on 23 March 2001 with a view to the adoption of a Directive of the European Parliament and of the Council on the promotion of electricity produced from renewable energy sources in the internal electricity market. No: 5583/1/01 (27 March 2001). Council of the European Union. (2001a). 2347th Council meeting: energy/industry. No: 8538/01 Presse: 181 (14-15 May 2001). Council of the European Union. (2002). Common position (EC) No 46/2002 adopted by the Council on 7 June 2002 with a view to the adoption of a Directive 2002/.../EC of the European Parliament and of the Council of ... on the energy performance of buildings. No: C 197 E/02 (20 August 2002). Council of the European Union. (2005a). Common position adopted by the Council on 8 December 2005 with a view to the adoption of a Decision of the European Parliament and of the Council laying down guidelines for trans- European energy networks and repealing Decisions Nos 96/391/EC and 1229/2003/EC. No: 10720/1/05 (8 December 2005). Council of the European Union. (2005b). Statement of the Council’s reasons. Common position adopted by the Council on 8 December 2005 with a view to the adoption of a Decision of the European Parliament and of the Council laying down guidelines for trans-European energy networks and repealing Decisions Nos 96/391/EC and 1229/2003/EC. No: 10720/1/05 (1 December 2005). Council of the European Union. (2008a). 2875th Council meeting: transport, telecommunications and energy. No: 10310/08 Presse: 162 (6 June 2008). Council of the European Union. (2008b). 2895th Council meeting: transport, telecommunications and energy. No: 13649/08 Presse: 276 (9 and 10 October 2008). Council of the European Union. (2008c). 2898th Council meeting: environment. No: 13857/08 Presse: 282 (20 October 2008). Council of the European Union. (2008d). 2913rd Council meeting: transport, telecommunications and energy. No: 16920/08 Presse: 262 (8 and 9 December 2008). Council of the European Union. (2008e). Preparation of the Council (environment) meeting on 5 June 2008 and of the TTE Council (energy) meeting on 6 June 2008. No: 9648/08 (26 May 2008).

306 Bibliography

Council of the European Union. (2009a). 2949th Council meeting: transport, telecommunications and energy. No: 10850/09 Presse: 169 (11 and 12 June 2009). Council of the European Union. (2009b). 2983rd meeting of the Council of the European Union (transport, telecommunications and energy). No: 17205/09 (6 January 2010). Council of the European Union. (2010). Position (EU) No 10/2010 of the Council at first reading with a view to the adoption of a Directive of the European Parliament and of the Council on the energy performance of buildings (recast). No: C 123 E/04 (12 May 2010). Council of the European Union. (2012). List of Council preparatory bodies. No: 12223/12 (4 July 2012). Crabbé, A., & Leroy, P. (2008). The handbook of environmental policy evaluation. London: Earthscan. Damro, C. (2012). Market power Europe. Journal of European Public Policy, 19(5), 682-699. Damro, C., Hardie, I., & MacKenzie, D. (2008). The EU and climate change policy: law, politics and prominence at different levels. Journal of Contemporary European Research, 4(3), 179-192. Dawson, B., & Spannagle, M. (2009). The complete guide to climate change. Oxon: Routledge. De Clerck-Sachsse, J., & Maciej Kaczynski, P. (2009). The European Parliament – more powerful, less legitimate? An outlook for the seventh term. CEPS working document, No. 314 (May 2009). Dehousse, R., & Thompson, A. (2012). Intergovernmentalists in the Commission: foxes in the henhouse? European Integration, 34(2), 113-132. Devuyst, Y. (2012). The constitutional and Lisbon treaties. In E. Jones, A. Menon & S. Weatherill (Eds.), The Oxford handbook of the European Union (pp. 163- 178). Oxford: Oxford University Press. Diez, T., & Wiener, A. (2009). Introducing the mosaic of integration theory: its past, present and future. In A. Wiener & T. Diez (Eds.), European integration theory (Second ed., pp. 1-22). Oxford: Oxford University Press. Dimitrov, R. S. (2010). Inside Copenhagen: the state of climate governance. Global Environmental Politics, 10(2), 18-24. Duffield, J. S., & Birchfield, V. L. (2011). Introduction: the recent upheaval in EU energy policy. In V. L. Birchfield & J. S. Duffield (Eds.), Towards a common European Union energy policy: problems, progress, and prospects (pp. 1-9). New York: Palgrave MacMillan.

307 Bibliography

Dupont, C. (2010). Political commitment to climate policy integration at EU level: the case of biodiversity policy. Edinburgh Europa Paper Series, 2010(5). Dupont, C. (2011). Climate policy integration in the EU. In W.L. Filho (Ed.), The economic, social and political elements of climate change (pp. 385-404). Berlin: Springer-Verlag. Dupont, C. & Oberthür, S. (2012). Insufficient climate policy integration in EU energy policy: the importance of the long-term perspective. Journal of Contemporary European Research, 8(2), 228-247. Dupont, C., & Primova, R. (2011). Combating complexity: the integration of EU climate and energy policies. European Integration Online Papers, 15 (Special mini-issue 1), Article 8. ECF. (2010). Roadmap 2050. A practical guide to a prosperous, low carbon Europe. Brussels: European Climate Foundation. Eckersley, R. (2012). Does leadership make a difference in international climate politics? Paper presented at the International Studies Association Annual Conference, San Diego, 2012. EEA. (2005). Environmental policy integration in Europe: state of play and evaluation framework. EEA report no 2/2005. Copenhagen: European Environment Agency. EEA. (2008). Energy and environment report 2008. Copenhagen: European Environment Agency. EEA. (2010). The European environment: state and outlook 2010 synthesis. Copenhagen: European Environment Agency. EEA. (2011). Renewable energy 2000 to 2010 – from toddler to teen. Copenhagen: European Environment Agency. EEA. (2012). Consumption and the environment – 2012 update. Copenhagen: European Environment Agency. Egeberg, M. (2010). The European Commission. In M. Cini & N. Pérez-Solórzano Borragán (Eds.), European Union politics (Third ed., pp. 125-140). Oxford: Oxford University Press. Eickhout, B., van den Born, G. J., Notenboom, J., van Oorschot, M., Ros, J. van Vuuren, D. & Westhoek, H. (2008). Local and global consequences of the EU renewable Directive for biofuels: testing the sustainability criteria. The Hague: PBL Netherlands Environmental Assessment Agency. Eikeland, P. O. (2011). EU internal energy market policy: achievements and hurdles. In V. L. Birchfield & J. S. Duffield (Eds.), Towards a common European Union energy policy: problems, progress, and prospects (pp. 13-40). New York: Palgrave MacMillan.

308 Bibliography

ENDS Europe. (1999a). EU Commission attacked over renewables law. 22 October 1999. ENDS Europe. (1999b). EU ministers request renewables framework. 12 May 1999. ENDS Europe. (1999c). Plan for EU renewable energy law shelved. 9 February 1999. ENDS Europe. (1999d). Strong EU renewable energy law demanded. 28 April 1999. ENDS Europe. (2000a). EU renewables law gets ministerial approval. 7 December 2000. ENDS Europe. (2000b). France aims for quick deal on EU renewables. 31 May 2000. ENDS Europe. (2000c). MEPs demand binding renewables targets. 17 November 2000. ENDS Europe. (2000d). Slow going for EU renewables Directive. 8 November 2000. ENDS Europe. (2001a). EU renewable energy law all but finalised. 20 June 2001. ENDS Europe. (2001b). MEPs defy ministers over buildings efficiency. 19 December 2001. ENDS Europe. (2001c). MEP stands by legal renewable energy targets. 6 March 2001. ENDS Europe. (2001d). Ministers neuter EU building efficiency drive. 5 December 2001. ENDS Europe. (2002). MEPs go soft on buildings energy efficiency. 13 September 2002. ENDS Europe. (2004). Insulation firms urge wider building energy law. 5 March 2004. ENDS Europe. (2005). Plea for broader EU buildings energy law. 16 March 2005. ENDS Europe. (2008a). Council agreement on EU climate plans in detail. 12 December 2008. ENDS Europe. (2008b). Council set to reject EU-wide renewables trading. 27 June 2008. ENDS Europe. (2008c). Crisis leaves the climate glass half empty. 19 December 2008. ENDS Europe. (2008d). EU governments’ renewables debate rumbles on. 30 July 2008. ENDS Europe. (2008e). EU renewable energy law deal confirmed. 9 December 2008. ENDS Europe. (2008f). EU states reach accord on renewables plan. 29 October 2008. ENDS Europe. (2008g). European leaders pass climate “credibility test”. 12 December 2008. ENDS Europe. (2008h). MEP to propose big changes to renewable plans. 13 May 2008.

309 Bibliography

ENDS Europe. (2008i). MEPs agree to compromise on EU renewables vote. 10 September 2008. ENDS Europe. (2008j). MEPs back beefed-up renewable energy law. 11 September 2008. ENDS Europe. (2008k). MEPs praise Sarkozy’s climate deal. 16 December 2008. ENDS Europe. (2008l). New EU renewable energy law “finalised”. 2 December 2008. ENDS Europe. (2008m). Rapporteur kicks off EU renewables law debate. 31 January 2008. ENDS Europe. (2008n). Revised building energy Directive proposed. 13 November 2008. ENDS Europe. (2009a). All new buildings to be “near-zero-energy” by 2020. 18 November 2009. ENDS Europe. (2009b). Council to clash with MEPs over efficient buildings. 2 June 2009. ENDS Europe. (2009c). EU building energy law revision “must go further”. 6 January 2009. ENDS Europe. (2009d). EU states raise doubts over greener building plans. 16 February 2009. ENDS Europe. (2009e). MEPs demand financing focus in buildings plan. 20 January 2009. ENDS Europe. (2009f). MEPs set 2019 deadline for zero-energy buildings. 23 April 2009. ENDS Europe. (2009g). States’ fears for green building law revision grow. 3 July 2009. ENDS Europe. (2010a). EU risks missing efficiency goal, report warns. 16 September 2010. ENDS Europe. (2010b). Financing for greener buildings “still an issue”. 29 February 2010. ENDS Europe. (2010c). MEPs want efficiency focus in EU energy plan. 29 June 2010. ENDS Europe. (2011a). Drop in energy demand boosted RE share in 2009. 18 February 2011. ENDS Europe. (2011b). €265m EU energy efficiency fund launched. 1 July 2011. EREC. (2010). RE-thinking 2050: a 100% renewable energy vision for the European Union. Brussels: European Renewable Energy Council.

310 Bibliography

EREC, & Greenpeace. (2010). Energy [r]evolution: Towards a fully renewable energy supply in the EU27. Brussels: European Renewable Energy Council and Greenpeace International. Eurelectric. (2010). Power choices. Pathways to carbon-neutral electricity in Europe by 2050. Brussels: Eurelectric. Eurobarometer. (2008). Europeans’ attitudes towards climate change. Brussels: European Commission. Eurobarometer. (2011). Climate change. Brussels: European Commission. Eurogas. (2011). Eurogas roadmap 2050. Brussels: Eurogas. European Commission. (1968). First guidelines for a Community energy policy. COM (68) 1040. European Commission. (1972). Necessary progress in community energy policy. COM (72) 1200. European Commission. (1973). Programme of environmental action of the European Communities. Part II: detailed description of the actions to be undertaken at Community level over the next two years. COM (73) 530 final C. European Commission. (1978). Energy objectives for 1990 and programmes of the member states. COM (78) 613 final. European Commission. (1979a). New lines of action by the European Community in the field of energy saving. COM (79) 312. European Commission. (1979b). Third report of the Community’s programme for energy saving. COM (79) 313. European Commission. (1984). Towards a European policy for the rational use of energy in the building sector. COM (84) 614. European Commission. (1988). Communication to the Council. The greenhouse effect and the Community. COM (88) 656. European Commission. (1990). Communication from the Commission: energy and the environment. COM (89) 369 final. European Commission. (1991). Proposal for a Council Directive on the indication by labelling and standard product information of the consumption of energy and other resources of household appliances. COM (1991) 285. European Commission. (1992a). Proposal for a Council Directive to limit carbon dioxide emissions by improving energy efficiency (SAVE programme). COM (92) 182. European Commission. (1992b). Specific actions for greater penetration for renewable energy sources: ALTENER. COM (92) 180.

311 Bibliography

European Commission. (1995). Green paper for a European Union energy policy. COM (94) 659. European Commission. (1996). Energy for the future: renewable sources of energy. Green paper for a Community Strategy. COM (1996) 576. European Commission. (1997a). Climate change – the EU approach for Kyoto. COM (97) 481. European Commission. (1997b). Energy for the future: renewable sources of energy. White paper for a Community strategy and action plan. COM (1997) 599. European Commission. (1998). Energy efficiency in the European Community – towards a strategy for the rational use of energy. COM (1998) 246. European Commission. (2000a). Action plan to improve energy efficiency in the European Community. COM (2000) 247. European Commission. (2000b). Amended proposal for a Directive of the European Parliament and of the Council on the promotion of electricity from renewable energy sources in the internal electricity market. COM (2000) 884. European Commission. (2000c). Communication from the Commission on EU policies and measures to reduce greenhouse gas emissions: towards a European Climate Change Programme (ECCP). COM (2000) 88 final. European Commission. (2000d). Green paper: towards a European strategy for the security of energy supply. COM (2000) 769 final. European Commission. (2000e). Proposal for a Directive of the European Parliament and of the Council on the promotion of electricity from renewable energy sources in the internal electricity market. COM (2000) 279. European Commission. (2001a). Communication from the Commission: withdrawal of Commission proposals which are no longer topical. COM (2001) 763 final/2. European Commission. (2001b). European governance: a white paper. COM (2001) 428. European Commission. (2001c). Proposal for a Decision of the European Parliament and of the Council amending Decision No 1254/96/EC laying down a series of guidelines for trans-European energy networks. COM (2001) 775. European Commission. (2001d). Proposal for a Directive of the European Parliament and of the Council on the energy performance of buildings. COM (2001) 226. European Commission. (2002a). Amended proposal for a Directive of the European Parliament and of the Council on the energy performance of buildings. COM (2002) 192. European Commission. (2002b). Consultation document: towards a reinforced culture of consultation and dialogue – proposal for general principles and minimum

312 Bibliography

standards for consultation of interested parties by the Commission. COM (2002) 277. European Commission. (2002c). Proposal for a Directive of the European Parliament and the Council concerning measures to safeguard security of natural gas supply. COM (2002) 488. European Commission. (2002d). Towards a reinforced culture of consultation and dialogue – general principles and minimum standards for consultation of interested parties by the Commission. COM (2002) 704. European Commission. (2003a). Commission staff working paper. Decision of the European Parliament and of the Council laying down guidelines for trans- European energy networks and repealing Decisions No 96/391/EC and No 1229/2003/EC. Extended impact assessment. SEC (2003) 1369. European Commission. (2003b). Proposal for a Decision of the European Parliament and of the Council laying down guidelines for trans-European energy networks and repealing Decisions No 96/391/EC and No 1229/2003/EC. COM (2003) 742 final. European Commission. (2004). Communication from the Commission. The share of renewable energy in the EU. Commission report in accordance with Article 3 of Directive 2001/77/EC, evaluation of the effect of legislative instruments and other Community policies on the development of the contribution of renewable energy sources in the EU and proposals for concrete actions. COM (2004) 366. European Commission. (2005a). Communication from the Commission to the European Parliament pursuant to the second subparagraph of Article 251(2) of the EC Treaty concerning the common position of the Council on the adoption of a Decision of the European Parliament and of the Council laying down guidelines for trans-European energy networks and repealing Decisions No 96/391/EC and No 1229/2003/EC. COM (2005) 716. European Commission. (2005b). Green paper on energy efficiency: doing more with less. COM (2005) 265 final. European Commission. (2006a). Action plan for energy efficiency: realising the potential. COM (2006) 545 final. European Commission. (2006b). Green paper: a European strategy for sustainable, competitive and secure energy. COM (2006) 105 final. European Commission. (2006c). Green paper: European transparency initiative. COM (2006) 194. European Commission. (2006d). Green paper follow-up action: report on progress in renewable energy. COM (2006) 849.

313 Bibliography

European Commission. (2006e). Opinion of the Commission pursuant to Article 251(2), third subparagraph, point (c) of the EC Treaty, on the European Parliament’s amendments to the Council’s common position regarding the proposal for a Decision of the European Parliament and of the Council laying down guidelines for trans-European energy networks and repealing Decisions No 96/391/EC and No 1229/2003/EC. COM (2006) 381. European Commission. (2006f). Summary report on the analysis of the debate on the green paper “a European strategy for sustainable, competitive and secure energy”. SEC (2006) 1500. European Commission. (2007a). Commission staff working document: accompanying document to the renewable energy road map. Renewable energies in the 21st century: building a more sustainable future. Impact assessment. SEC (2006) 1719. European Commission. (2007b). Communication from the Commission: renewable energy road map. Renewable energies in the 21st century: building a more sustainable future. COM (2006) 848. European Commission. (2007c). Energy for a changing world – an energy policy for Europe. COM (2007) 1. European Commission. (2007d). Follow-up to the green paper “European transparency initiative”. COM (2007) 127. European Commission. (2007e). Limiting global climate change to 2 degrees Celsius. The way ahead for 2020 and beyond. COM (2007) 2. European Commission. (2008a). 20 20 by 2020: Europe’s climate change opportunity. COM (2008) 30. European Commission. (2008b). Accompanying document to the proposal for a recast of the energy performance of buildings Directive (2002/91/EC). Impact assessment. SEC (2008) 2864. European Commission. (2008c). Communication on the Directive 2004/67/EC of 26 April 2004 concerning measures to safeguard security of natural gas supply. COM (2008) 769. European Commission. (2008d). Proposal for a Directive of the European Parliament and of the Council on the promotion of the use of energy from renewable sources. COM (2008) 19. European Commission. (2008e). Second strategic energy review: an EU energy security and solidarity action plan. COM (2008) 781. European Commission. (2009a). Adapting to climate change: towards a European framework for action. COM (2009) 147. European Commission. (2009b). European transparency initiative: the register of interest representatives one year after. COM (2009) 612.

314 Bibliography

European Commission. (2009c). Proposal for a Regulation of the European Parliament and of the Council concerning measures to safeguard security of gas supply and repealing Directive 2004/67/EC. COM (2009) 363. European Commission. (2009d). Proposal for a Regulation of the European Parliament and of the Council establishing a programme to aid economic recovery by granting Community financial assistance to projects in the field of energy. COM (2009) 35. European Commission. (2010a). Country file: Libya. Brussels: Market Observatory for Energy. European Commission. (2010b). Country file: Russia. Brussels: Market Observatory for Energy. European Commission. (2010c). Country file: Ukraine. Brussels: Market Observatory for Energy. European Commission. (2010d). Energy infrastructure priorities for 2020 and beyond – a blueprint for an integrated European energy network. COM (2010) 677. European Commission. (2010e). Proposal for a Regulation of the European Parliament and of the Council amending Regulation (EC) No 663/2009 establishing a programme to aid economic recovery by granting Community financial assistance to projects in the field of energy. COM (2010) 283. European Commission. (2010f). Report from the Commission to the Council and the European Parliament on the implementation of the European energy programme for recovery. COM (2010) 191. European Commission. (2011a). Communication from the Commission: energy roadmap 2050. COM (2011) 885/2. European Commission. (2011b). Communication from the Commission. A roadmap for moving to a competitive low carbon economy in 2050. COM (2011) 112/4. European Commission. (2011c). Country file: Algeria. Brussels: Market Observatory for Energy. European Commission. (2011d). Country file: Norway. Brussels: Market Observatory for Energy. European Commission. (2011e). Executive summary of the impact assessment accompanying the document: energy roadmap 2050. SEC (2011) 1566. European Commission. (2011f). Impact assessment accompanying the document: energy roadmap 2050. SEC (2011) 1565 part one. European Commission. (2011g). Impact assessment accompanying the document: energy roadmap 2050. SEC (2011) 1565 part two. European Commission. (2011h). Impact assessment accompanying the document: proposal for a Regulation of the European Parliament and of the Council on

315 Bibliography

guidelines for trans-European energy infrastructure and repealing Decision No 1364/2006/EC. SEC (2011) 1233. European Commission. (2011i). Report from the Commission to the Council and the European Parliament on the implementation of the European energy programme for recovery. COM (2011) 217(20 April 2011). European Commission. (2012a). Country file: Turkey. Brussels: Market Observatory for Energy. European Commission. (2012b). Report from the Commission to the European Parliament and the Council on the implementation of the European energy programme for recovery. COM (2012) 445. European Commission. (2013a). An EU strategy on adaptation to climate change. COM (2013) 216. European Commission. (2013b). Commission staff working document accompanying the document: financial support for energy efficiency in buildings. SWD (2013) 143. European Commission. (2013c). Report from the Commission to the European Parliament and the Council: financial support for energy efficiency in buildings. COM (2013) 225. European Commission. (2013d). Report from the Commission to the European Parliament and the Council: renewable energy progress report. COM (2013) 175. European Council. (1988). Annex I: declaration on the environment, 2-3 December 1988. European Council. (1990). Presidency conclusions, 25-26 June 1990. European Council. (1996). Presidency conclusions, 21-22 June 1996. European Council. (2006). Presidency conclusions, 23-24 March 2006. European Council. (2007). Presidency conclusions, 8-9 March 2007. European Council. (2008a). Presidency conclusions, 13-14 March 2008. European Council. (2008b). Presidency conclusions, 11-12 December 2008. European Council. (2009a). Presidency conclusions, 19-20 March 2009. European Council. (2009b). Presidency conclusions, 29-30 October 2009. European Parliament. (2000a). European Parliament first reading on the proposal for a European Parliament and Council Directive on the promotion of electricity from renewable energy sources in the internal electricity market. T5- 0514/2000 (16 November 2000).

316 Bibliography

European Parliament. (2000b). Report on the proposal for a European Parliament and Council Directive on the promotion of electricity from renewable sources in the internal electricity market. A5-0320/2000 (30 October 2000). European Parliament. (2001a). Recommendation for second reading on the Council common position for adopting a European Parliament and Council Directive on the promotion of electricity produced from renewable energy sources in the internal electricity market. A5-0227/2001 (22 June 2001). European Parliament. (2001b). Report on the proposal for a Directive of the European Parliament and of the Council on the energy performance of buildings. A5- 0465/2001 (19 December 2001). European Parliament. (2002a). Position of the European Parliament adopted at first reading on 6 February 2002 with a view to the adoption of European Parliament and Council Directive .../.../EC on the energy performance of buildings. T5-0039/2002 (6 February 2002). European Parliament. (2002b). Recommendation for second reading on the Council common position with a view to the adoption of a Directive of the European Parliament and of the Council on the energy performance of buildings. A5- 0297/2002 (15 September 2002). European Parliament. (2005). Report on the proposal for a Decision of the European Parliament and of the Council laying down guidelines for trans-European energy networks and repealing Decisions No 96/391/EC and No 1229/2003/EC. A6-0134/2005 (3 May 2005). European Parliament. (2006a). European Parliament legislative resolution on the Council common position for adopting a Decision of the European Parliament and of the Council laying down guidelines for trans-European energy networks and repealing Decision No 96/391/EC and Decision No 1229/2003/EC. T6-0118/2006 (4 April 2006). European Parliament. (2006b). European Parliament resolution on a European strategy for sustainable, competitive and secure energy – green paper. T6- 0603/2006 (14 December 2006). European Parliament. (2009a). European Parliament legislative resolution of 23 April 2009 on the proposal for a Directive of the European Parliament and of the Council on the energy performance of buildings (recast; first reading). T6- 0278/2009 (23 April 2009). European Parliament. (2009b). Gas and oil pipelines in Europe. Brussels: European Parliament, Directorate-General for internal policies. Policy department A: economic and scientific policy. European Parliament. (2009c). Report on the proposal for a Directive of the European Parliament and of the Council on the energy performance of buildings (recast; first reading). A6-0254/2009 (1 April 2009).

317 Bibliography

European Parliament. (2010). Opinion of the Committee on the environment, public health and food safety for the Committee on industry, research and energy on the proposal for a Regulation of the European Parliament and Council concerning measures to safeguard security of gas supply and repealing Directive 2004/67/EC. A7-0112/2010 (29 January 2010). European Parliament. (2012). Co-decision and conciliation: a guide to how the Parliament co-legislates under the Treaty of Lisbon. Brussels: European Parliament, Directorate-General for internal policies of the Union and Directorate for legislative coordination and conciliations. EWEA. (2011). EU energy policy to 2050: achieving 80-95% emissions reductions. Brussels: European Wind Energy Association. ExxonMobil. (2012). 2012, the outlook for energy: a view to 2040. Irving, TX: ExxonMobil. Falkner, R., Stephan, H., & Vogler, J. (2010). International climate policy after Copenhagen: towards a “building blocks” approach. Global Policy, 1(3), 252- 262. Fioretos, O. (2011). Historical institutionalism and international relations. International Organization, 65(2), 367-399. Friedrichs, J. & Kratochwil, F. (2009). On acting and knowing: how pragmatism can advance international relations research and methodology. International Organization, 63(4), 701-731. Gas LNG Europe. (2011). LNG Map. Brussels: Gas Infrastructure Europe. Geller, H., Harrington, P., Rosenfeld, A. H., Tanishima, S., & Unander, F. (2006). Policies for increasing energy efficiency: thirty years of experience in OECD countries. Energy Policy, 34(5), 556-573. George, A. L., & Bennett, A. (2004). Case studies and theory development in the social sciences. London, England and Cambridge, MA: MIT Press. Gerring, J. (2007). Case study research: principles and practices. Cambridge: Cambridge University Press. Gerring, J. (2008). Case selection for case-analysis: qualitative and quantitative techniques. In J. M. Box-Steffensmeier, H. E. Brady & D. Collier (Eds.), The Oxford handbook of political methodology (pp. 645-684). Oxford: Oxford University Press. Giddens, A. (2009). The politics of climate change. Cambridge: Polity Press. Giordano, V., Gangale, F., Fulli, G., & Sanchez Jimenez, M. (2011). Smart grid projects in Europe: lessons learned and current developments. Brussels: Joint Research Centre.

318 Bibliography

Gornitzka, Å., & Sverdrup, U. (2008). Who consults? The configuration of expert groups in the European Union. West European Politics, 31(4), 725-750. Greenpeace. (2013). Why is Europe burning more coal? Retrieved 6 June 2013 from http://www.greenpeace.org.uk/newsdesk/energy/analysis/why-europe-burning- more-coal. Gupta, J., & Grubb, M. (Eds.). (2000). Climate change and European leadership. Dordrecht: Kluwer Academic Publishers. Haas, E. B. (1958). The uniting of Europe: political, social and economic forces, 1950-1957. Stanford, CA: Stanford University Press. Haas, E. B. (1961). International integration: the European and universal process. International Organization, 15(3), 366-392. Haas, E. B. (1970). The study of regional integration: reflections on the joy and anguish of pretheorizing. International Organization, 24(4), 607-646. Haas, E. B. (2001). Does constructivism subsume neofunctionalism? In T. Christiansen, K. E. Jørgensen & A. Wiener (Eds.), The social construction of Europe (pp. 22-31). London: Sage Publications Ltd. Hadfield, A. (2008). EU-Russia energy relations: aggregation and aggravation. Journal of Contemporary European Studies, 16(2), 231-248. Hall, P. A., & Taylor, R. C. R. (1996). Political science and the three new institutionalisms. Political Studies, 44(5), 936-957. Harvey, F. (2012). Gas rebranded as green energy by EU. The Guardian. Retrieved 29 April 2013 from http://www.guardian.co.uk/environmenta/2012/may/29/gas- rebranded-green-energy-eu?INTCMP=SRCH. Haug, C., & Jordan, A. (2010). Burden sharing: distributing burdens or sharing efforts? In A. Jordan, D. Huitema, H. van Asselt, T. Rayner & F. Berkhout (Eds.), Climate change policy in the European Union: confronting the dilemmas of adaptation and mitigation? (pp. 83-102). Cambridge: Cambridge University Press. Hauser, H. (2011). European Union lobbying post Lisbon: an economic analysis. Berkeley Journal of International Law, 29(2), 680-709. Haverland, M. & Romeijn, M. (2007). Do member states make European policies work? Analysing the EU transposition deficit. Public Administration, 85(3), 757-778. Heaps, C., Erickson, P., Kartha, S., & Kemp-Benedict, E. (2009). Europe’s share of the climate challenge: domestic actions and international obligations to protect the planet. Stockholm: Stockholm Environment Institute.

319 Bibliography

Henningsen, J. (2011). Energy savings and efficiency. In V. L. Birchfield & J. S. Duffield (Eds.), Towards a common European Union energy policy: problems, progress, and prospects (pp. 131-141). New York: Palgrave MacMillan. Hermanns, H. (2004). Interviewing as an activity. In U. Flick, E. von Kardoff & I. Steinke (Eds.), A companion to qualitative research (pp. 209-213). London: Sage Publications Ltd. Herodes, M., Adelle, C., & Pallemaerts, M. (2007). Environmental policy integration at the EU level: a literature review. EPIGOV Paper No. 5. Ecologic – Institute for International and European Environmental Policy. Hildingsson, R., Stripple, J., & Jordan, A. (2010). Renewable energies: a continuing balancing act? In A. Jordan, D. Huitema, H. van Asselt, T. Rayner & F. Berkhout (Eds.), Climate change policy in the European Union: confronting the dilemmas of mitigation and adaptation? (pp. 103-124). Cambridge: Cambridge University Press. Hix, S., & Høyland, B. (2011). The political system of the European Union (Third ed.). Houndmills: Palgrave MacMillan. Hoffmann, S. (1966). Obstinate of obsolete? The fate of the nation-state and the case of Western Europe. Daedalus, 95(5), 862-915. Hopf, C. (2004). Qualitative interviews: an overview. In U. Flick, E. von Kardoff & I. Steinke (Eds.), A companion to qualitative research (pp. 203-208). London: Sage Publications Ltd. Howarth, R. W., Santoro, R., & Ingraffea, A. (2011). Methane and the greenhouse-gas footprint of natural gas from shale formations: a letter. Climatic Change, 106(4), 679-690. Howarth, R. W., Santoro, R., & Ingraffea, A. (2012). Venting and leaking of methane from shale gas development: response to Cathles et al. Climatic Change, 113(2), 537-549. Howes, T. (2010). The EU’s new renewable energy Directive (2009/28/EC). In S. Oberthür & M. Pallemaerts (Eds.), The new climate policies of the European Union: internal legislation and climate diplomacy (pp. 117-150). Brussels: VUB Press. IEA. (2011a). World energy outlook 2011. Paris: OECD/International Energy Agency. IEA. (2011b). World energy outlook 2011: are we entering a golden age of gas? Paris: OECD/International Energy Agency. IEA. (2012). Natural gas information. Paris: OECD/International Energy Agency. IEA. (2013). Tracking clean energy progress 2013: IEA input to the clean energy ministerial. Paris: OECD/International Energy Agency.

320 Bibliography

IPCC. (1995). Second assessment report of the Intergovernmental Panel on Climate Change. Cambridge: Cambridge University Press. IPCC. (2007). Climate change 2007. Fourth assessment report: synthesis report. Geneva: Intergovernmental Panel on Climate Change. Jacob, K., & Hertin, J. (2007). Evaluating integrated impact assessments – a conceptual framework. EPIGOV Paper No. 7. Ecologic – Institute for International and European Environmental Policy. Jacob, K., Volkery, A., & Lenschow, A. (2008). Instruments for environmental policy integration in 30 OECD countries. In A. Jordan & A. Lenschow (Eds.), Innovation in environmental policy? Integrating the environment for sustainability (pp. 24-45). Cheltenham: Edward Elgar. Jansen, J. C., & Uyterlinde, M. A. (2004). A fragmented market on the way to harmonisation? EU policy-making on renewable energy promotion. Energy for Sustainable Development, 8(1), 93-107. Jones, C. (2010). The European Union’s policy on the development of renewable energy. In B. Moselle, J. Padilla & R. Schmalensee (Eds.), Harnessing renewable energy in electric power systems: theory, practice, policy. London: Earthscan. Jordan, A. (2002). Efficient hardware and light green software: environmental policy integration in the UK. In A. Lenschow (Ed.), Environmental policy integration: greening sectoral policies in Europe (pp. 35-56). London: Earthscan. Jordan, A., Huitema, D., & van Asselt, H. (2010). Climate change policy in the European Union: an introduction. In A. Jordan, D. Huitema, H. van Asselt, T. Rayner & F. Berkhout (Eds.), Climate change policy in the European Union: confronting the dilemmas of mitigation and adaptation? (pp. 3-25). Cambridge: Cambridge University Press. Jordan, A., Huitema, D., van Asselt, H., Rayner, T., & Berkhout, F. (Eds.). (2010). Climate change policy in the European Union: confronting the dilemmas of mitigation and adaptation? Cambridge: Cambridge University Press. Jordan, A., & Lenschow, A. (2008a). Environmental policy integration: an innovation in environmental policy? In A. Jordan & A. Lenschow (Eds.), Innovation in environmental policy? Integrating the environment for sustainability (pp. 313- 341). Cheltenham: Edward Elgar Publishing Ltd. Jordan, A., & Lenschow, A. (Eds.). (2008b). Innovation in environmental policy? Integrating the environment for sustainability. Cheltenham: Edward Elgar Publishing Ltd. Jordan, A., & Lenschow, A. (2008c). Integrating the environment for sustainable development: an introduction. In A. Jordan & A. Lenschow (Eds.), Innovation

321 Bibliography

in environmental policy? Integrating the environment for sustainability (pp. 3- 23). Cheltenham: Edward Elgar Publishing Ltd. Jordan, A., & Lenschow, A. (2010). Environmental policy integration: a state of the art review. Environmental Policy and Governance, 20(3), 147-158. Jordan, A., & Rayner, T. (2010). The evolution of climate policy in the European Union: an historical overview. In A. Jordan, D. Huitema, H. van Asselt, T. Rayner & F. Berkhout (Eds.), Climate change policy in the European Union: confronting the dilemmas of mitigation and adaptation? (pp. 52-80). Cambridge: Cambridge University Press. Jordan, A., Schout, A., & Unfried, M. (2008). The European Union. In A. Jordan & A. Lenschow (Eds.), Innovation in environmental policy? Integrating the environment for sustainability (pp. 159-179). Cheltenham: Edward Elgar. Keohane, R. O., & Nye, J. (2011). Power and independence: world politics in transition (Fourth ed.). New York: Longman Publishing Group. King, G., Keohane, R. O., & Verba, S. (1994). Designing social inquiry: scientific inference in qualitative research. Princeton, N.J.: Princeton University Press. Kingdon, J. W. (2003). Agendas, alternatives, and public policies (Second ed.). London: Longman. Kivimaa, P., & Mickwitz, P. (2006). The challenge of greening technologies: environmental policy integration in Finnish technology policies. Research Policy, 35(5), 729-744. Klotz, A. (2008). Case selection. In A. Klotz & D. Prakesh (Eds.), Qualitative methods in international relations: a pluralist guide (pp. 43-58). New York: Palgrave Macmillan. Kulovesi, K., Morgera, E., & Muñoz, M. (2010). The EU’s climate and energy package: environmental integration and international dimensions. Edinburgh Europa Paper Series, 2010(38). Kvale, S., & Brinkman, S. (2009). Interviews: learning the craft of qualitative research interviewing (Second ed.). London: Sage Publications Ltd. Lacasta, N., Oberthür, S., Santos, E., & Barata, P. (2010). From sharing the burden to sharing the effort: Decision 406/2009/EC on member state emission targets for non-ETS sectors. In S. Oberthür & M. Pallemaerts (Eds.), The new climate policies of the European Union: internal legislation and climate diplomacy (pp. 93-116). Brussels: VUB Press. Lafferty, W. M., & Hovden, E. (2003). Environmental policy integration: towards an analytical framework. Environmental Politics, 12(5), 1-22. Lafferty, W. M., & Knudsen, J. (2007). The issue of “balance” and trade-offs in environmental policy integration: how will we know EPI when we see it?

322 Bibliography

EPIGOV paper no. 11. Ecologic – Institute for International and European Environmental Policy. Lampinen, R. & Uusikylä, P. (1998). Implementation deficit – why member states do not comply with EU Directives. Scandinavian Political Studies, 21(3), 231- 251. Larsson, R. L. (2007). Nord Stream, Sweden and Baltic Sea security. Stockholm: FOI – Defence Research Agency. Lauber, V. (2001). The different concepts of promoting RES-electricity and their political careers. Paper presented at the 2001 Berlin Conference on the Human Dimensions of Global Environmental Change, “Global environmental change and the nation state”, Berlin. Lelieveldt, H., & Princen, S. (2011). The politics of the European Union. Cambridge: Cambridge University Press. Lenschow, A. (2002a). Greening the European Union: an introduction. In A. Lenschow (Ed.), Environmental policy integration: greening sectoral policies in Europe (pp. 3-21). London: Earthscan. Lenschow, A. (Ed.). (2002b). Environmental policy integration: greening sectoral policies in Europe. London: Earthscan. Lenschow, A. (2010). Environmental policy: contending dynamics of policy change. In H. Wallace, M. A. Pollack & A. R. Young (Eds.), Policy-making in the European Union (Sixth ed., pp. 307-330). Oxford: Oxford University Press. Liberatore, A. (1997). The integration of sustainable development objectives into EU policymaking. In S. Baker, M. Kousis, D. Richardson & S. Young (Eds.), The politics of sustainable development (pp. 107-126). London: Routledge. Liefferink, D., Arts, B., Kamstra, J., & Ooijevaar, J. (2009). Leaders and laggards in environmental policy: a quantitative analysis of domestic policy outputs. Journal of European Public Policy, 16(5), 677-700. Lindberg, L. N. (1963). The political dynamics of European economic integration. Palo Alto, CA: Stanford University Press. Lustgarten, A. (2011). Climate benefits of natural gas may be overstated. Scientific American. Retrieved 15 April 2013 from http://www.scientificamerican.com/article.cfm?id=climate-benefits-natural- gas-overstated. Manners, I. (2002). Normative power Europe: a contradiction in terms? Journal of Common Market Studies, 40(2), 235-258. Manners, I. (2006). Normative power Europe reconsidered: beyond the crossroads. Journal of European Public Policy, 13(2), 182-199.

323 Bibliography

March, J. G., & Olsen, J. P. (1989). Rediscovering institutions: the organizational basis of politics. New York: The Free Press. McCormick, J. (2005). Understanding the European Union (Third ed.). Houndmills: Palgrave Macmillan. Mercier, A., & Peteves, S. D. (2006). Results of the public consultation on the EU initiative on heating and cooling from renewable energy sources. Brussels: Joint Research Centre. Metcalfe, L. (1994). International policy co-ordination and public management reform. International Review of Administrative Sciences, 60(2), 271-290. Mickwitz, P., Aix, F., Beck, S., Carss, D., Ferrand, N., Görg, C., Jensen, A., Kivimaa, P., Kuhlicke, C., Kuindersma, W., Máñez, M., Melanen, M., Monni, S., Pedersen, A. B., Reinert, H., & van Bommel, S. (2009). Climate policy integration, coherence and governance. PEER Report No. 2. Helsinki: Partnership for European Environmental Research. Moomaw, W., Yamba, F., Kamimoto, M., Maurice, L., Nyboer, J., Urama, K., & Weir, T. (2011). Introduction. In O. Edenhofer, R. Pichs-Madruga, Y. Sokona, K. Seyboth, P. Matschoss, S. Kadner, T. Zwickel, P. Eickemeier, G. Hansen, S. Schlömer & C. von Stechow (Eds.), IPCC special report on renewable energy sources and climate change mitigation. Cambridge: Cambridge University Press. Moravcsik, A. (1993). Preferences and power in the European Community: a liberal intergovernmentalist approach. Journal of Common Market Studies, 31(4), 473-524. Moravcsik, A. (1998). The choice for Europe. Social purpose and state power from Messina to Maastricht. Oxon: Routledge. Moravcsik, A., & Schimmelfennig, F. (2009). Liberal intergovernmentalism. In A. Wiener & T. Diez (Eds.), European integration theory (Second ed., pp. 67-87). Oxford: Oxford University Press. Niemann, A., & Schmitter, P. C. (2009). Neofunctionalism. In A. Wiener & T. Diez (Eds.), European integration theory (Second ed., pp. 45-66). Oxford: Oxford University Press. Nilsson, M. (2011). EU renewable electricity policy: mixed emotions towards harmonization. In V. L. Birchfield & J. S. Duffield (Eds.), Towards a common European Union energy policy: problems, progress, and prospects (pp. 113- 130). New York: Palgrave MacMillan. Nilsson, M., & Eckerberg, K. (Eds.). (2007). Environmental policy integration in practice: shaping institutions for learning. London: Earthscan. Nilsson, M., Eckerberg, K., Hagberg, L., Swartling, Å. G., & Söderberg, C. (2007). Policy framing and EPI in energy and agriculture. In M. Nilsson & K.

324 Bibliography

Eckerberg (Eds.), Environmental policy integration in practice: shaping institutions for learning (pp. 85-110). London: Earthscan. Nilsson, M., & Nilsson, L. J. (2005). Towards climate policy integration in the EU: evolving dilemmas and opportunities. Climate Policy, 5(3), 363-376. Nilsson, M., Nilsson, L. J. & Ericsson, K. (2009). The rise and fall of GO trading in European renewable energy policy: the role of advocacy and policy framing. Energy Policy, 37(11), 4454-4462. Nilsson, M., & Persson, Å. (2003). Framework for analysing environmental policy integration. Journal of Environmental Policy and Planning, 5(4), 333-359. Nohrstedt, D. (2006). External shocks and policy change: Three Mile Island and Swedish nuclear energy policy. Journal of European Public Policy, 12(6), 1041-1059. Nollkamper, A. (2002). Three conceptions of the integration principle in international environmental law. In A. Lenschow (Ed.), Environmental policy integration: greening sectoral policies in Europe (pp. 22-34). London: Earthscan. Oberthür, S. (2007). The European Union in international climate policy: the prospect for leadership. Intereconomics – Review of European Economic Policy 42(2), 77-83. Oberthür, S. (2009). Interplay management: enhancing environmental policy integration among international institutions. International Environmental Agreements: Politics, Law and Economics, 9(4), 371-391. Oberthür, S. (2011). The European Union’s performance in the international climate change regime. Journal of European Integration 33(6), 667-682. Oberthür, S., & Dupont, C. (2011). The Council, the European Council and international climate policy: from symbolic leadership to leadership by example. In R. K. W. Wurzel & J. Connelly (Eds.), The European Union as a leader in international climate change politics (pp. 74-91). London and New York: Routledge. Oberthür, S., & Pallemaerts, M. (2010a). The EU’s internal and external climate policies: an historical overview. In S. Oberthür & M. Pallemaerts (Eds.), The new climate policies of the European Union: internal legislation and climate diplomacy (pp. 27-63). Brussels: VUB Press. Oberthür, S., & Pallemaerts, M. (Eds.). (2010b). The new climate policies of the European Union: internal legislation and climate diplomacy. Brussels: VUB Press. Oberthür, S., & Roche Kelly, C. (2008). EU leadership in international climate policy: achievements and challenges. International Spectator, 43(3), 35-50.

325 Bibliography

OECD (2002). Improving policy coherence and integration for sustainable development: a checklist. Paris: Organisation for Economic Co-operation and Development. Pallemaerts, M., & Williams, R. (2006). Climate change: the international and European policy framework. In M. Peeters & K. Deketelaere (Eds.), EU climate change policy. The challenge of new regulatory initiatives (pp. 22-50). Cheltenham: Edward Elgar. Persson, Å. (2004). Environmental policy integration: an introduction. PINTS – Policy integration for sustainability, background paper. Stockholm: Stockholm Environment Institute. Persson, Å. (2007). Different perspectives on EPI. In M. Nilsson & K. Eckerberg (Eds.), Environmental policy integration in practice: shaping institutions for learning (pp. 25-48). London: Earthscan. Peters, B. G. (1998). Managing horizontal government: the politics of co-ordination. Public Administration, 76(2), 295-311. Peters, B. G. (2011). Institutional theory in political science: the new institutionalism. London: Continuum Publishing Corporation. Pierson, P. (1998). The path to European integration: a historical-institutionalist analysis. In W. Sandholtz & A. Stone Sweet (Eds.), European integration and supranational governance (pp. 27-58). Oxford: Oxford University Press. Piket, V. (2003). EU/Russia energy dialogue – state of progress. Paper presented at the Sixth Annual General Meeting of the European Business Congress, Varna, Bulgaria. Pirani, S. (2012). Russo-Ukrainian gas wars and the call on transit governance. In C. Kuzemko, A. V. Belyi, A. Goldthau & M. F. Keating (Eds.), Dynamics of energy governance in Europe and Russia (pp. 169-186). Houndmills: Palgrave Macmillan. Pirani, S., Stern, J., & Yafimava, K. (2009). The Russo-Ukrainian gas dispute of January 2009: a comprehensive assessment. Oxford Institute for Energy Studies, NG 27 (February 2009). Piris, J.-C. (2010). The Lisbon Treaty: a legal and political analysis. Cambridge: Cambridge University Press. Pointvogl, A. (2009). Perceptions, realities, concession – what is driving the integration of European energy policies? Energy Policy, 37(12), 5704-5716. Pollack, M. A. (1998). The engines of integration? Supranational autonomy and influence in the European Union. In W. Sandholtz & A. Stone Sweet (Eds.), European integration and supranational governance (pp. 217-249). Oxford: Oxford University Press.

326 Bibliography

Pollack, M. A. (2009). The new institutionalisms and European integration. In A. Wiener & T. Diez (Eds.), European integration theory (Second ed., pp. 125- 143). Oxford: Oxford University Press. Pollack, M. A. (2010). Theorizing EU policy-making. In H. Wallace, M. A. Pollack & A. R. Young (Eds.), Policy-making in the European Union (Sixth ed., pp. 15- 44). Oxford: Oxford University Press. Pollack, M. A., & Hafner-Burton, E. M. (2010). Mainstreaming international governance: the environment, gender and IO performance in the European Union. The Review of International Organizations, 5(3), 285-313. PricewaterhouseCoopers. (2010). 100% renewable electricity. A roadmap to 2050 for Europe and North Africa. UK: PricewaterhouseCoopers, pwc.com/uk. Rasmussen, M. K. (2012). Is the European Parliament still a policy champion for environmental issues? Interest Groups & Advocacy, 1(2), 239-259. Rathbun, B. C. (2008). Interviewing and qualitative field methods: pragmatism and practicalities. In J. M. Box-Steffensmeier, H. E. Brady & D. Collier (Eds.), The Oxford handbook of political methodology (pp. 685-701). Oxford: Oxford University Press. Reichardt, K., Pfluger, B., Schleich, J., & Marth, H. (2012). With or without CCS? Decarbonising the EU power sector. Responses Policy Update, 3 (July 2012). Richter, B. (2010). Beyond smoke and mirrors: climate change and energy in the 21st century. Cambridge: Cambridge University Press. Ries, C. P., Jenkins, J., & Wise, O. (2009). Improving the energy performance of buildings: learning from the European Union and Australia. Santa Monica, CA: RAND Environment, Energy, and Economic Development. Rietig, K. (2012). Climate policy integration beyond principled priority: a framework for analysis. Centre for Climate Change Economics and Policy, Working Paper No. 99. Risse, T. (2005). Neofunctionalism, European identity, and the puzzles of European integration. Journal of European Public Policy, 12(2), 291-309. Risse, T. (2009). Social constructivism and European integration. In A. Wiener & T. Diez (Eds.), European integration theory (Second ed., pp. 144-160). Oxford: Oxford University Press. Risse-Kappen, T. (1996). Exploring the nature of the beast: international relations theory and comparative policy analysis meet the European Union. Journal of Common Market Studies, 34(1), 53-80. Rohlfing, I. (2012). Case studies and causal inference: an integrative framework. Houndmills: Palgrave Macmillan.

327 Bibliography

Roosa, S. A. (2010). Sustainable development handbook. Lilburn, GA: The Fairmont Press. Rosamond, B. (2005). The uniting of Europe and the foundation of EU studies: revisiting the neofunctionalism of Ernst B. Haas. Journal of European Public Policy, 12(2), 237-254. Rosamond, B. (2007). New theories of European integration. In M. Cini (Ed.), European Union politics (Second ed., pp. 117-136). Oxford: Oxford University Press. Rowlands, I. H. (2005). The European Directive on renewable electricity: conflicts and compromises. Energy Policy, 33(8), 965-974. Sabatier, P. A. (Ed.). (2007). Theories of the policy process (Second ed.). Boulder: Westview Press. Scharpf, F. W. (1988). The joint-decision trap: lessons from German federalism and European integration. Public Administration, 66(3), 239-278. Scharpf, F. W. (2006). The joint-decision trap revisited. Journal of Common Market Studies, 44(4), 845-864. Schout, A., & Jordan, A. (2005). Coordinated European governance: self-organizing or centrally steered? Public Administration, 83(1), 201-220. Schout, A., & Jordan, A. (2008a). Administrative instruments. In A. Jordan & A. Lenschow (Eds.), Innovation in environmental policy? Integrating the environment for sustainability (pp. 49-69). Cheltenham: Edward Elgar Publishing Limited. Schout, A., & Jordan, A. (2008b). EU-EPI, Policy co-ordination and new institutionalism. Institut Universitari d’Estudis Europeus. Schreurs, M. A., & Tiberghien, Y. (2007). Multi-level reinforcement: explaining European Union leadership in climate change mitigation. Global Environmental Politics, 7(4), 19-46. Shell International BV. (2008). Shell energy scenarios to 2050. The Hague: Shell International BV. Skea, J., Chaudry, M. & Wang, X. (2012). The role of gas infrastructure in promoting UK energy security. Energy Policy, 43, 202-213. Skjærseth, J. B. (1994). The climate policy of the EC: too hot to handle? Journal of Common Market Studies, 32(1), 25-45. Skjærseth, J. B. & Wettestad, J. (2008). EU emissions trading: initiation, decision- making and implementation. Aldershot: Ashgate. Stephenson, E., Doukas, A., & Shaw, K. (2012). Greenwashing gas: might a “transition fuel” label legitimize carbon-intensive natural gas development? Energy Policy, 46(July 2012), 452-459.

328 Bibliography

Stern, N. (2007). The economics of climate change: the Stern review. Cambridge: Cambridge University Press. Steurer, R. (2007). From government strategies to strategic public management: an exploratory outlook on the pursuit of cross-sectoral policy integration. European Environment, 17(3), 201-214. Stone Sweet, A., & Sandholtz, W. (1998). Integration, supranational governance, and the institutionalization of the European polity. In W. Sandholtz & A. Stone Sweet (Eds.), European integration and supranational governance (pp. 1-26). Oxford: Oxford University Press. Strøby-Jensen, C. (2007). Neo-functionalism. In M. Cini (Ed.), European Union politics (Second ed., pp. 85-98). Oxford: Oxford University Press. Swanborn, P. (2010). Case study research: what, why and how? London: Sage Publications Ltd. Tallberg, J. (2004). The power of the Presidency: brokerage, efficiency and distribution in EU negotiations. Journal of Common Market Studies, 42(5), 999-1022. Tanasescu, I. (2009). The European Commission and interest groups: towards a deliberative interpretation of stakeholder involvement in EU policy-making. Brussels: VUB Press. Tsebelis, G., & Garrett, G. (2001). The institutional foundations of intergovernmentalism and supranationalism in the European Union. International Organization, 55(2), 357-390. Underdal, A. (1980). Integrated marine policy: what? Why? How? Marine Policy, 4(3), 159-169. van Asselt, H. (2010). Emissions trading: the enthusiastic adoption of an “alien” instrument? In A. Jordan, D. Huitema, H. van Asselt, T. Rayner & F. Berkhout (Eds.), Climate change policy in the European Union: confronting the dilemmas of adaptation and mitigation? (pp. 125-144). Cambridge: Cambridge University Press. van Bommel, S., & Kuindersma, W. (2008). Policy integration, coherence and governance in Dutch climate policy: a multi-level analysis of mitigation and adaptation policy. Alterra-rapport 1799. Wageningen: Alterra. Vasileiadou, E., & Tuinstra, W. (2012). Stakeholder consultations: mainstreaming climate policy in the energy directorate? Environmental Politics, 22(3), 475- 495. von Hirschhausen, C. (2010). Developing a supergrid. In B. Moselle, J. Padilla & R. Schmalensee (Eds.), Harnessing renewable energy in electric power systems: theory, practice, policy (pp. 181-206). London: Earthscan.

329 Bibliography von Homeyer, I. (2006). Environmental policy integration and modes of governance – state-of-the-art report. EPIGOV Paper No. 2. Berlin: Ecologic – Institute for International and European Environmental Policy. Wallace, H., Pollack, M. A., & Young, A. R. (Eds.). (2010). Policy-making in the European Union (Sixth ed.). Oxford: Oxford University Press. Warleigh-Lack, A., & Drachenberg, R. (2010). Policy making in the European Union. In M. Cini & N. Pérez-Solórzano Borragán (Eds.), European Union politics (Third ed., pp. 209-224). Oxford: Oxford University Press. Warntjen, A. (2007). Steering the Union. The impact of the EU Presidency on legislative activity in the Council. Journal of Common Market Studies, 45(5), 1135-1157. Watson, R., & Corbett, R. (2012). How policies are made. In E. Bomberg, J. Peterson & R. Corbett (Eds.), The European Union: how does it work? (Third ed., pp. 122-140). Oxford: Oxford University Press. Watson, R., & Shackleton, M. (2003). Organized interests and lobbying in the EU. In E. Bomberg & A. Stubb (Eds.), The European Union: how does it work? (pp. 88-107). Oxford: Oxford University Press. Wesselink, B., Harmsen, R., & Eichhammer, W. (2010). Energy savings 2020: how to triple the impact of energy saving policies in Europe. Brussels: European Climate Foundation. Wettestad, J. (2000). The complicated development of EU climate policy. In J. Gupta & M. Grubb (Eds.), Climate change and European leadership: A sustainable role for Europe? (pp. 25-45). Dordrecht: Kluwer Academic Publishers. Wettestad, J. (2005). The making of the 2003 EU emissions trading Directive: an ultra-quick process due to entrepreneurial proficiency? Global Environmental Politics, 5(1), 1-23. Wettestad, J., Eikeland, P. O., & Nilsson, M. (2012). EU climate and energy policy: a hesitant supranational turn? Global Environmental Politics, 12(2), 67-86. Wiener, A., & Diez, T. (2009a). Taking stock of integration theory. In A. Wiener & T. Diez (Eds.), European integration theory (Second ed., pp. 241-252). Oxford: Oxford University Press. Wiener, A., & Diez, T. (Eds.). (2009b). European integration theory (Second ed.). Oxford: Oxford University Press. Wigley, T. M. L. (2011). Coal to gas: the influence of methane leakage. Climatic Change, 108(4), 601-608. Wilkinson, D. (1998). Steps towards integrating the environment into other EU policy sectors. In T. O’Riordan & H. Voisey (Eds.), The transition to sustainability: the politics of agenda 21 in Europe (pp. 113-129). London: Earthscan.

330 Bibliography

Wilkinson, D. (2007). Environmental policy integration at EU level – state-of-the-art report. EPIGOV Paper No 4. Berlin: Ecologic, Institute for International and European Environmental Policy. Wurzel, R. K. W., & Connelly, J. (2011a). Introduction: European Union political leadership in international climate change politics. In R. K. W. Wurzel & J. Connelly (Eds.), The European Union as a leader in international climate change politics (pp. 3-20). London: Routledge. Wurzel, R. K. W., & Connelly, J. (Eds.). (2011b). The European Union as a leader in international climate change politics. London: Routledge. WWF. (2007). Climate solutions. WWF’s vision for 2050. Gland: World Wide Fund for Nature. WWF. (2011). The energy report: 100% renewable energy by 2050. Gland: World Wide Fund for Nature. Yin, R. K. (2003). Applications of case study research (Second ed.). London: Sage Publications Ltd. Yin, R. K. (2009). Case study research: Design and methods (Fourth ed.). London: Sage Publications Ltd. Zingerli, C., Bisang, K., & Zimmermann, W. (2004). Towards policy integration: experiences with intersectoral coordination in international and national forest policy. Paper presented at the Berlin Conference 2004 on the Human Dimension of Global Environmental Change “Greening of Policies – Interlinkages and Policy Integration”.

331

332

Annex I

List of legislation

Council Decision 91/565/EEC of 29 October 1991 concerning the promotion of energy efficiency in the Community (SAVE programme).

Council Decision 93/389/EEC of 24 June 1993 for a monitoring mechanism of

Community CO2 and other greenhouse gas emissions.

Council Decision 93/500/EEC of 13 September 1993 concerning the promotion of renewable energy sources in the Community (ALTENER programme).

Council Directive 89/106/EEC of 21 December 1988 on the approximation of laws, regulations and administrative provisions of the Member States relating to construction products.

Council Directive 92/42/EEC of 21 May 1992 on efficiency requirements for new hot-water boilers fired with liquid or gaseous fuels.

Council Directive 92/775/EEC of 22 September 1992 on the indication by labelling and standard product information of the consumption of energy and other resources by household appliances.

Council Directive 93/76/EEC of 13 September 1993 to limit carbon dioxide emissions by improving energy efficiency (SAVE).

Council Directive 96/391/EC of 28 March 1996 laying down a series of measures aimed at creating a more favourable context for the development of trans- European networks in the energy sector.

Council Directive 2004/67/EC of 26 April 2004 concerning measures to safeguard security of natural gas supply.

Council Regulation (EC) No 2236/95 of 18 September 1995 laying down general rules for the granting of Community financial aid in the field of trans- European networks.

333

Decision No 1254/96/EC of the European Parliament and of the Council of 5 June 1996 laying down a series of guidelines for trans-European energy networks.

Decision No 1229/2003/EC of the European Parliament and of the Council of 26 June laying down a series of guidelines for trans-European energy networks and repealing Decision No 1254/96/EC.

Decision No 280/2004/EC of the European Parliament and of the Council of 11 February 2004 concerning a mechanism for monitoring Community greenhouse gas emissions and for implementing the Kyoto Protocol.

Decision No 1364/2006/EC of the European Parliament and of the Council of 6 September 2006 laying down guidelines for trans-European energy networks and repealing Decision No 96/391/EC and Decision No 1229/2003/EC.

Decision No 406/2009/EC of the European Parliament and of the Council of 23 April 2009 on the effort of Member States to reduce their greenhouse gas emissions to meet the Community’s greenhouse gas emission reduction commitments up to 2020.

Directive 96/92/EC of the European Parliament and of the Council of 19 December 1996 concerning common rules for the internal market in electricity.

Directive 98/30/EC of the European Parliament and of the Council of 22 June 1998 concerning common rules for the internal market in natural gas.

Directive 2001/77/EC of the European Parliament and of the Council of 27 September 2001 on the promotion of electricity produced from renewable energy sources in the internal electricity market.

Directive 2002/91/EC of the European Parliament and of the Council of 16 December 2002 on the energy performance of buildings.

Directive 2003/54/EC of the European Parliament and of the Council of 26 June 2003 concerning common rules for the internal market in electricity and repealing Directive 96/92/EC.

334

Directive 2003/55/EC of the European Parliament and of the Council of 26 June 2003 concerning common rules for the internal market in natural gas and repealing Directive 98/30/EC.

Directive 2003/87/EC of the European Parliament and of the Council of 13 October 2003 establishing a scheme for greenhouse gas emission allowance trading within the Community and amending Council Directive 96/61/EC.

Directive 2004/8/EC of the European Parliament and of the Council of 11 February 2004 on the promotion of cogeneration based on a useful heat demand in the internal energy market and amending Directive 92/42/EEC.

Directive 2004/101/EC of the European Parliament and of the Council of 27 October 2004 amending Directive 2003/87/EC establishing a scheme for greenhouse gas emission allowance trading within the Community, in respect of the Kyoto Protocol’s project mechanisms.

Directive 2005/32/EC of the European Parliament and of the Council of 6 July 2005 establishing a framework for the setting of ecodesign requirements for energy- using products and amending Council Directive 92/42/EEC and Directives 96/57/EC and 2000/55/EC of the European Parliament and of the Council.

Directive 2006/32/EC of the European Parliament and of the Council of 5 April 2006 on energy end-use efficiency and energy services and repealing Council Directive 93/76/EEC.

Directive 2006/40/EC of the European Parliament and of the Council of 17 May 2006 relating to emissions from air conditioning systems in motor vehicles and amending Council Directive 70/156/EEC.

Directive 2009/28/EC of the European Parliament and of the Council of 23 April 2009 on the promotion of the use of energy from renewable sources and amending and subsequently repealing Directives 2001/77/EC and 2003/30/EC.

Directive 2009/29/EC of the European Parliament and of the Council of 23 April 2009 amending Directive 2003/87/EC so as to improve and extend the greenhouse gas emission allowance trading scheme of the Community.

335

Directive 2009/31/EC of the European Parliament and of the Council of 23 April 2009 on the geological storage of carbon dioxide and amending Council Directive 85/337/EEC, European Parliament and Council Directives 2000/60/EC, 2004/35/EC, 2008/1/EC and Regulation (EC) No 1013/2006.

Directive 2009/125/EC of the European Parliament and of the Council of 21 October 2009 establishing a framework for the setting of ecodesign requirements for energy-related products (recast).

Directive 2010/30/EU of the European Parliament and of the Council of 19 May 2010 on the indication by labelling and standard product information of the consumption of energy and other resources by energy-related products (recast).

Directive 2010/31/EU of the European Parliament and of the Council of 19 May 2010 on the energy performance of buildings (recast).

Directive 2012/27/EU of the European Parliament and of the Council of 25 October 2012 on energy efficiency, amending Directives 2009/125/EU and 2010/30/EU and repealing Directives 2004/8/EC and 2006/32/EC.

Regulation (EC) No 663/2009 of the European Parliament and of the Council of 13 July 2009 establishing a programme to aid economic recovery by granting Community financial assistance to projects in the field of energy.

Regulation (EU) No 994/2010 of the European Parliament and of the Council of 20 October 2010 concerning measures to safeguard security of gas supply and repealing Council Directive 2004/67/EC.

Regulation (EU) No 211/2011 of the European Parliament and of the Council of 16 February 2011 on the citizens’ initiative.

336

Annex II

Legal texts

Treaty establishing the European Community (TEC): …

Article 6

Environmental protection requirements must be integrated into the definition and implementation of the Community policies and activities referred to in Article 3, in particular with a view to promoting sustainable development.

…

Article 95

1. By way of derogation from Article 94 and save where otherwise provided in this Treaty, the following provisions shall apply for the achievement of the objectives set out in Article 14. The Council shall, acting in accordance with the procedure referred to in Article 251 and after consulting the Economic and Social Committee, adopt the measures for the approximation of the provisions laid down by law, regulation or administrative action in Member States which have as their object the establishment and functioning of the internal market.

2. Paragraph 1 shall not apply to fiscal provisions, to those relating to the free movement of persons nor to those relating to the rights and interests of employed persons.

3. The Commission, in its proposals envisaged in paragraph 1 concerning health, safety, environmental protection and consumer protection, will take as a base a high level of protection, taking account in particular of any new development based on scientific facts. Within their respective powers, the European Parliament and the Council will also seek to achieve this objective.

337

4. If, after the adoption by the Council or by the Commission of a harmonisation measure, a Member State deems it necessary to maintain national provisions on grounds of major needs referred to in Article 30, or relating to the protection of the environment or the working environment, it shall notify the Commission of these provisions as well as the grounds for maintaining them.

5. Moreover, without prejudice to paragraph 4, if, after the adoption by the Council or by the Commission of a harmonisation measure, a Member State deems it necessary to introduce national provisions based on new scientific evidence relating to the protection of the environment or the working environment on grounds of a problem specific to that Member State arising after the adoption of the harmonisation measure, it shall notify the Commission of the envisaged provisions as well as the grounds for introducing them.

6. The Commission shall, within six months of the notifications as referred to in paragraphs 4 and 5, approve or reject the national provisions involved after having verified whether or not they are a means of arbitrary discrimination or a disguised restriction on trade between Member States and whether or not they shall constitute an obstacle to the functioning of the internal market.

In the absence of a decision by the Commission within this period the national provisions referred to in paragraphs 4 and 5 shall be deemed to have been approved.

When justified by the complexity of the matter and in the absence of danger for human health, the Commission may notify the Member State concerned that the period referred to in this paragraph may be extended for a further period of up to six months.

7. When, pursuant to paragraph 6, a Member State is authorised to maintain or introduce national provisions derogating from a harmonisation measure, the Commission shall immediately examine whether to propose an adaptation to that measure.

8. When a Member State raises a specific problem on public health in a field which has been the subject of prior harmonisation measures, it shall bring it to the attention of the Commission which shall immediately examine whether to propose appropriate measures to the Council.

338

9. By way of derogation from the procedure laid down in Articles 226 and 227, the Commission and any Member State may bring the matter directly before the Court of Justice if it considers that another Member State is making improper use of the powers provided for in this Article.

10. The harmonisation measures referred to above shall, in appropriate cases, include a safeguard clause authorising the Member States to take, for one or more of the non- economic reasons referred to in Article 30, provisional measures subject to a Community control procedure.

…

Article 175

1. The Council, acting in accordance with the procedure referred to in Article 251 and after consulting the Economic and Social Committee and the Committee of the Regions, shall decide what action is to be taken by the Community in order to achieve the objectives referred to in Article 174.

2. By way of derogation from the decision-making procedure provided for in paragraph 1 and without prejudice to Article 95, the Council, acting unanimously on a proposal from the Commission and after consulting the European Parliament, the Economic and Social Committee and the Committee of the Regions, shall adopt:

- provisions primarily of a fiscal nature;

- measures concerning town and country planning, land use with the exception of waste management and measures of a general nature, and management of water resources;

- measures significantly affecting a Member State’s choice between different energy sources and the general structure of its energy supply.

The Council may, under the conditions laid down in the preceding subparagraph, define those matters referred to in this paragraph on which decisions are to be taken by a qualified majority.

3. In other areas, general action programmes setting out priority objectives to be attained shall be adopted by the Council, acting in accordance with the procedure

339

referred to in Article 251 and after consulting the Economic and Social Committee and the Committee of the Regions.

The Council, acting under the terms of paragraph 1 or paragraph 2 according to the case, shall adopt the measures necessary for the implementation of these programmes.

4. Without prejudice to certain measures of a Community nature, the Member States shall finance and implement the environment policy.

5. Without prejudice to the principle that the polluter should pay, if a measure based on the provisions of paragraph 1 involves costs deemed disproportionate for the public authorities of a Member State, the Council shall, in the act adopting that measure, lay down appropriate provisions in the form of:

- temporary derogations, and/or

- financial support from the Cohesion Fund set up pursuant to Article 161.

…

Treaty on the functioning of the European Union (TFEU): …

Article 4

1. The Union shall share competence with the Member States where the Treaties confer on it a competence which does not relate to the areas referred to in Articles 3 and 6.

2. Shared competence between the Union and the Member States applies in the following principal areas:

(a) internal market;

(b) social policy, for the aspects defined in this Treaty;

(d) agriculture and fisheries, excluding the conservation of marine biological resources;

340

(e) environment;

(f) consumer protection;

(g) transport;

(h) trans-European networks;

(i) energy;

(j) area of freedom, security and justice;

(k) common safety concerns in public health matters, for the aspects defined in this Treaty.

3. In the areas of research, technological development and space, the Union shall have competence to carry out activities, in particular to define and implement programmes; however, the exercise of that competence shall not result in Member States being prevented from exercising theirs.

4. In the areas of development cooperation and humanitarian aid, the Union shall have competence to carry out activities and conduct a common policy; however, the exercise of that competence shall not result in Member States being prevented from exercising theirs.

…

Article 11

(ex Article 6 TEC)

Environmental protection requirements must be integrated into the definition and implementation of the Union’s policies and activities, in particular with a view to promoting sustainable development.

…

TITLE XVI

TRANS-EUROPEAN NETWORKS

Article 170

341

(ex Article 154 TEC)

1. To help achieve the objectives referred to in Articles 26 and 174 and to enable citizens of the Union, economic operators and regional and local communities to derive full benefit from the setting-up of an area without internal frontiers, the Union shall contribute to the establishment and development of trans-European networks in the areas of transport, telecommunications and energy infrastructures.

2. Within the framework of a system of open and competitive markets, action by the Union shall aim at promoting the interconnection and interoperability of national networks as well as access to such networks. It shall take account in particular of the need to link island, landlocked and peripheral regions with the central regions of the Union.

Article 171

(ex Article 155 TEC)

1. In order to achieve the objectives referred to in Article 170, the Union:

- shall establish a series of guidelines covering the objectives, priorities and broad lines of measures envisaged in the sphere of trans-European networks; these guidelines shall identify projects of common interest,

- shall implement any measures that may prove necessary to ensure the interoperability of the networks, in particular in the field of technical standardisation,

- may support projects of common interest supported by Member States, which are identified in the framework of the guidelines referred to in the first indent, particularly through feasibility studies, loan guarantees or interest-rate subsidies; the Union may also contribute, through the Cohesion Fund set up pursuant to Article 177, to the financing of specific projects in Member States in the area of transport infrastructure.

The Union’s activities shall take into account the potential economic viability of the projects.

2. Member States shall, in liaison with the Commission, coordinate among themselves the policies pursued at national level which may have a significant impact on the achievement of the objectives referred to in Article 170. The Commission may, in

342

close cooperation with the Member States, take any useful initiative to promote such coordination.

3. The Union may decide to cooperate with third countries to promote projects of mutual interest and to ensure the interoperability of networks.

Article 172

(ex Article 156 TEC)

The guidelines and other measures referred to in Article 171(1) shall be adopted by the European Parliament and the Council, acting in accordance with the ordinary legislative procedure and after consulting the Economic and Social Committee and the Committee of the Regions.

Guidelines and projects of common interest which relate to the territory of a Member State shall require the approval of the Member State concerned.

…

TITLE XX

ENVIRONMENT

Article 191

(ex Article 174 TEC)

1. Union policy on the environment shall contribute to pursuit of the following objectives:

- preserving, protecting and improving the quality of the environment,

- protecting human health,

- prudent and rational utilisation of natural resources,

- promoting measures at international level to deal with regional or worldwide environmental problems, and in particular combating climate change.

…

343

TITLE XXI

ENERGY

Article 194

1. In the context of the establishment and functioning of the internal market and with regard for the need to preserve and improve the environment, Union policy on energy shall aim, in a spirit of solidarity between Member States, to:

(a) ensure the functioning of the energy market;

(b) ensure security of energy supply in the Union;

(d) promote the interconnection of energy networks.

2. Without prejudice to the application of other provisions of the Treaties, the European Parliament and the Council, acting in accordance with the ordinary legislative procedure, shall establish the measures necessary to achieve the objectives of paragraph 1. Such measures shall be adopted after consultation of the Economic and Social Committee and the Committee of the Regions.

Such measures shall not affect a Member State’s right to determine the conditions for exploiting its energy resources, its choice between different energy sources and the general structure of its energy supply, without prejudice to Article 192(2)(c).

3. By way of derogation from paragraph 2, the Council, acting in accordance with a special legislative procedure, shall unanimously and after consulting the European Parliament, establish the measures referred to therein when they are primarily of a fiscal nature.

…

Article 293

(ex Article 250 TEC)

344

1. Where, pursuant to the Treaties, the Council acts on a proposal from the Commission, it may amend that proposal only by acting unanimously, except in the cases referred to in paragraphs 10 and 13 of Article 294, in Articles 310, 312 and 314 and in the second paragraph of Article 315.

2. As long as the Council has not acted, the Commission may alter its proposal at any time during the procedures leading to the adoption of a Union act.

…

Rules of procedure of the European Parliament …

Rule 50

Where a question of competence is referred to the Conference of Presidents pursuant to Rules 188(2) or 48, and the Conference of Presidents, on the basis of Annex VII, considers that the matter falls almost equally within the competence of two or more committees, or that different parts of the matter fall within the competence of two or more committees, Rule 49 shall apply with the following additional provisions:

- the timetable shall be jointly agreed by the committees concerned; - the rapporteurs concerned shall keep each other informed and shall endeavour to agree on the texts they propose to their committees and on their position regarding amendments; - the chairs and rapporteurs concerned shall jointly identify areas of the text falling within their exclusive or joint competence and agree on the precise arrangements for their cooperation. In the event of disagreement about the delimitation of competences the matter shall be submitted, at the request of one of the committees involved, to the Conference of Presidents, which may decide on the question of the respective competences or decide that the procedure with joint committee meetings under Rule 51 is to apply; the second subparagraph of Rule 188(2) shall apply mutatis mutandis; - the committee responsible shall accept without a vote amendments from an associated committee where they concern matters which fall within the exclusive competence of the associated committee. If amendments on matters

345

which fall within the joint competence of the committee responsible and an associated committee are rejected by the former, the latter may table those amendments directly in plenary; - in the event of a conciliation procedure taking place on the proposal, Parliament’s delegation shall include the rapporteur of any associated committee.

346

Annex III

List of interviews

No. Interviewee Date 1. European renewable industry representative 26 January 2010 2. DG Climate Action official 06 July 2010 3. DG Climate Action official 06 July 2010 4. Member of the Scottish Parliament 03 August 2010 5. European climate NGO representative 10 January 2012 6. European environmental NGO representative 09 February 2012 7. European environmental NGO representative 15 February 2012 7 May 2012 & 20 March 8. European environmental NGO representative 2013 9. DG Energy official 25 June 2012 10. European energy efficiency industry representative 05 July 2012 11. European energy efficiency industry representative 12 July 2012 12. DG Energy official 01 August 2012 13. Gas industry representative 10 September 2012 14. Estonian environmental NGO representative 14 September 2012 15. Estonian environmental NGO representative 17 September 2012 16. Estonian environmental NGO representative 18 September 2012 17. Austrian environmental NGO representative 18 September 2012 18. Latvian environmental NGO representative 18 & 21 September 2012 19. DG Energy official 03 October 2012 20. European climate research institute representative 24 October 2012 21. Energy market analyst 23 January 2013 22. Former DG Energy official 06 February 2013 23. DG Climate Action official 15 March 2013 24. DG Energy official 20 March 2013 25. Member of the European Parliament 27 March 2013 26. Member of the European Parliament 09 April 2013 *Five of the above interviews were carried out by email.

347