Transcript Boegbeeld Benchmarking

Performance based emissions trading post 2012
IIR Conference “CO2 in the Industry”
Carlton Oasis Hotel Spijkenisse, Netherlands,
19-20 September 2007
Jan Berends
Manager Environment & Product Safety
Royal DSM
[email protected]
Vianney Schyns
Manager Climate & Energy Efficiency
Utility Support Group
Utility provider for a.o. DSM and SABIC
[email protected]
Contents
1. Present political situation
2. Failures present EU ETS rules – examples at
DSM & SABIC
3. Key aspects of benchmarks
4. Benchmarks ex-ante or ex-post
5. Setting the total cap 2013-2020 – effectiveness
6. Conclusion
7. Annexes
2
Present political situation (1)
• Historical grandfathering was a historical mistake
= Recognised by EU Commission
• 3rd Trading period: perhaps auctioning for electricity & (partial?)
auctioning and/or benchmarking for industry
• EU Commission will come with a proposal December 2007 – then
co-decision EU Parliament & Council
– Takes 1.5 – 2 years, is no decision for single Member State
• Benchmarking for allocation to operators
= Ex-ante: based on historical production
= Ex-post: based on actual production
3
Present political situation (2)
• Industry is against auctioning
– Auctioning electricity
= Electricity prices remain high  bad for competitiveness / leakage
= Windfall profits nuclear, hydro (EU 45%)  what about France (90%)?
= Revenue recycling poses problems with effectiveness of EU ETS
– Auctioning industry
= Bad for competitiveness, “leakage” by production relocation
– Auctioning with Border Tax Adjustments at EU borders is neither
practical (huge bureaucracy) nor politically desired
4
Failures present
rules of the
EU Emissions Trading Scheme
Examples DSM & SABIC
Chemelot examples of ineffective allocation
• Incumbent plants
– Efficiency factor ß (total energy worldtop / total energy actual) of all
plants and all energy use/CO2-emissions is related only to local
emission of few plants
 Incentive CO2 reduction largely taken away  SEE ANNEX for
example Chemelot
• New entrants (new plants or extensions existing plants)
– Dutch rule = “never more allowances than needed”, zero incentive
for investments in cleaner technology
 New DSM melamine plant: 70% better than Best Practice
 Previous design work new SABIC large scale plant
6
Key aspects of benchmarks
Few benchmarks – high coverage
Suitable benchmark formula
Benchmarks take account of all energy carriers
Benchmarks in a direct emission scheme
Same effectiveness as auctioning
Examples chemical industry
Few benchmarks – major coverage, Pareto
Benchmarking Netherlands: about 90
100%
Coverage
of
emissions
under the
EU ETS
Major chemicals (20-30)
Refineries (1)
Cement (2)
Policy recommendation:
include (co-)firing biomass
Steel (6-7)
Electricity (1) and
for CHP (Combined Heat
& Power) (1 for heat)
Allocation:
• Vital few: about 40 benchmarks
• Trivial many: basis is own
efficiency, “be generous”,
give incentive to reduce
emissions
8
Suitable benchmark formula
Benchmark = WAE – CF x {WAE – BP}
9
Benchmark takes account of all energy carriers
Many energy functions can
be done either with:
• Steam, or
• Electricity, or
• Natural gas or other fuel
CO2 ?
Production
plant
Feeds
Steam
Product(s)
Electricity
Natural gas ?
Other fuel ?
Benchmark takes this into
account:
Normalised calculation to (total)
primary energy – or total CO2
Benchmark for only fuel –
direct emissions – is
meaningless
Examples: chemical plants, refineries, cement, paper plants, etc.
10
Benchmarks in a direct emissions scheme
Allocation = direct emission – emission {total plant – total BM}
CO2
Example 1:
• Net-import of secondary energy carriers:
Production
plant
Feeds
Product(s)
70 – {120 – 100} = 50
Plant worse than benchmark
Example 2:
Steam
Electricity
Natural gas
Other fuels
70 – {90 – 100} = 80
Plant better than benchmark
See formula in ANNEX
Site utilities have also benchmarks
11
Benchmarking same effectiveness as auctioning (1)
 Incentive to reduce emissions is independent of the exact
value of benchmark in a certain year
Incentive = avoided purchases + sales of allowances
Example:
Investment to reduce emissions from 900 to 600 kg CO2 per unit
of product (in old plant or new plant)
• Year 1, BM = 750: incentive = 150 + 150 = 300
• Year n, BM = 700: incentive = 200 + 100 = 300
 Predictability of investment climate
12
Benchmarking same effectiveness as auctioning (2)
 Same benchmarks for incumbents and new plants
Recognised by EU Commission
 Avoids
= Distorting transfer rules
= Barriers to entry
= Enhanced market concentration
 Ensures
= Equal incentive for plant improvement & plant replacement
 No “maximisation” or “minimisation” rules, see e.g. Matthes
(NL 110% and 85% now)
13
Examples chemical industry prove it works
EU bechmark data major chemicals
Product
Consultant
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
Weighted
EU Best
Efficiencies
PSR = WAE - CF x (WAE - BP)
EU average Practice
Electricity Heat
CF = Compliance Factor =
WAE
BP
15%
20%
GJ/ton
GJ/ton
Steamcrackers (1)
Solomon Associates
144,8
107,8
37,5%
90%
139,3
137,4
Pyrolosis gasoline (pygas)
Process Design Centre
1,3
0,6
42%
90%
1,2
1,2
Benzene extraction
Process Design Centre
3,8
2,2
42%
90%
3,6
3,5
Butadiene
Solomon Associates
9,72
7,3
37,5%
90%
9,4
9,2
MTBE
Process Design Centre
1,9
1,06
42%
90%
1,8
1,7
ldPE (low density polyethylene)
Phillip Townsend Associates
8,53
5,96
42%
90%
8,1
8,0
hdPE (high density polyethylene)
Phillip Townsend Associates
5,43
3,14
42%
90%
5,1
5,0
PP (polypropylene)
Phillip Townsend Associates
3,56
2,27
42%
90%
3,4
3,3
EPDM (ethylene propylene rubber) (2) Phillip Townsend Associates
32,22
28,0
42%
90%
31,6
31,4
PVC (polyvinyl chloride)
Process Design Centre
3,8
3,4
42%
90%
3,7
3,7
Nylon-6
Process Design Centre
10,0
5,71
42%
90%
9,4
9,1
Ammonia (3)
Plant Services International
13,13
7,23
40%
90%
12,2
11,9
Nitric acid
Process Design Centre
-0,12
-1,8
42%
90%
-0,4
-0,5
Fertiliser (Calcium Ammonium Nitrate)Process Design Centre
0,99
0,35
42%
90%
0,9
0,9
Urea
Plant Services International
5,06
3,06
42%
90%
4,8
4,7
Melamine (4)
Nexant
79,46
60,55
42%
90%
76,6
75,7
Caprolactam excl. cyclohexanon
Process Design Centre
8,7
-0,9
42%
90%
7,3
6,8
Acrylonitril (2)
Phillip Townsend Associates
-6,2
-8,3
42%
90%
-6,5
-6,6
Yeast
Process Design Centre
5,9
5,62
42%
90%
5,9
5,8
1) Solomon energy efficiency index (EEI) adjusted for supplemental feeds
2) WAE and BP are not EU but worldwide data (for confidentialilty reasons)
3) 20.67 GJ/ton feedstock energy (these process emissions fall outside the EU ETS)
4) These data include feedstock use which must be subtracted: 29.5 GJ/ton ammonia and 21.99 GJ/ton urea incl. ammonia use.
Typicals are: 3.2 ton urea and -0.9 ton ammonia, both per ton melamine. This gives WAE = 35.6 GJ/ton melamine and BP = 16.7 GJ/ton melamine.
Once defined, benchmarks are quite straightforward
14
Benchmarks ex-ante or ex-post
Quality of historic production data & forecasts
Guarantee of total cap
Solution of “production subsidy”
Historic production tells nothing about the future
Quality of historic data for operators
… with climate change instruments based on history?
Variations in annual load
factors over five years, found in
UK by consultant NERA
for UK government
Link to actual production:
 Avoids distortions
 Avoids windfall profits
 Solves problems new entrants
and closures, SEE ANNEX
16
Quality of historic data & forecasts for states
• What means a historic cap: many new plants enter the market?
– Many new power plants in Italy around 2009 .. Germany .. NL
• What means a historic cap: import or export of product?
– More electricity import NL from Germany – Is NL then doing well?
– New CHP in Luxembourg – Is Luxembourg doing bad?
• What means a historic cap: economy is strongly recovering?
– Forecast of growth in central Europe, seven legal cases European Court
of Justice against EU Commission: Czech Republic, Estonia, Hungary,
Latvia, Poland, Slovakia, Lithuania
– Influence Burden Sharing on allocation is perverse
• Solution: benchmarks linked to actual production
17
Benchmark with ex-post + guarantee total cap
Benchmark with ex-post electricity
(without contingency reserve)
FORECASTS
Start
Ex-post
over 2008
done in 2009
to 2010
Ex-post
over 2012
done in 2013
to 2014
Scenario with a higher production growth than forecasted
Second trading period
Third period
2008
2009
2010
2011
2012
Total
2013
Production fossil, TWh
2000
2034
2069
2104
2140
10346
Benchmark, ton CO2/MWh
0,600
0,590
0,580
0,570
0,561
Total cap, Mton CO2
1200
1200
1200
1200
1200
6000
Fixed
Fixed
Update production fossil, TWh
Ex-post, TWh
Ex-post, Mton
Allocation, Mton CO2
Benchmark, ton CO2/MWh
Total cap, Mton CO2
Update production fossil, TWh
Ex-post, TWh
Ex-post, Mton
Allocation, Mton CO2
Benchmark, ton CO2/MWh
Total cap, Mton CO2
2030
2034
1200
0,600
1200
Fixed
1200
0,590
1200
Fixed
2030
2045
1200
0,600
1200
Fixed
1200
0,590
1200
Fixed
2090
30
18
1194
0,571
1212
Fixed
2130
30
18
1194
0,571
1212
Fixed
2125
2155
1194
0,562
1194
1194
0,554
1194
2140
11
6
1191
0,563
1197
Fixed
2175
40
23
1168
0,538
1191
Fixed
2014
10434 Update forecast
6000
10520
6000
2190
25
14
986
0,450
1011
Fixed
2230
5
3
997
0,447
1002
Fixed
• Automatic adjustments within an ex-ante agreed total cap
• More stringent benchmarks work exactly like auctioning (& cap & trade)
• System is self-adjusting; virtually no interest costs
18
Benchmarking in the product chain
Benchmarking with ex-post adjustment to actual production provides
incentives in the product chain … avoids “production subsidy” effect
(= higher electricity demand by lower electricity prices)
Electricity
Fuel
Electricity and
heat
generation
Fuel
Feed
Industrial
manufacturing plant
with use of
electricity and heat
Product
Heat, from
CHP or from
boilers
… the efficiency of
the production of
electricity & heat
… the efficiency of
the use of (fuel),
electricity & heat
19
Setting the total cap 2013-2020
while maintaining an effective CO2-price
Ex-post cap maybe too stringent: exploding CO2-price
Ex-post cap maybe too soft: collapse CO2-price
What is the solution?
Setting total cap 2013-2020 – effectiveness
• Total cap may appear to be too stringent
– Renewables behind target, delay phase-out German nuclear, higher
economic growth than expected, etc.  very high CO2-price
• Total cap may appear to be too soft
– Reverse of possible causes above very low CO2-price
 Ex-ante frozen allocation not effective
• Solution
– Contingency reserve if cap too stringent  for example 100 Mton
– No loser benchmarks if cap too soft
 Target is to maintain a realistic CO2-price to ensure a robust and
predictable EU ETS for companies to reduce emissions
21
Conclusion
 Performance-based allocation
-
Can be realized with guarantee of the total cap
Creates same incentives for emission reduction as auctioning
Can be the basis for a stable & gradually increasing CO2-price
22
Transition for a global trading scheme
Benchmark:
Specific
energy use
or CO2
emission
Benchmark China-India
Incentive low carbon technologies
the same in global trading scheme
Benchmark USA-Canada
Benchmark EU-Japan
Global benchmark
Transition period (with 3 or more BMs for same product) avoids
high cost in case of auctioning for regions with higher emissions per
unit of product (vital: BMs without differentiation new/old plants)
2008
2012
2017
2022
2027
2032
23
Annexes
What’s wrong with present EU ETS rules
Benchmarking 2008-2012 to Chemelot incumbent plants
Permit site Chemelot = C + D + E
CO2
CO2
CO2
A. External power
Plants in ETS Electricity
B. Combined
heat & power
Swentibold in ETS
C. Utility production
site Chemelot
“EdeA” in ETS
D. Plants outside ETS 2008-2012:
ammonia, nitric acid, fertilisers,
melamine, polymers, etc.
CO2
CO2
E. Plants in ETS 2008-2012:
two steam crackers
Steam
Allocation permit site Chemelot = ßD + E x emission E + ßC x emission C
while for example: emission C could be too high, although ßC = 1.0
Ineffective allocation – lack of incentive – no true benchmarking
25
Benchmarking 2008-2012 to Chemelot incumbent plants
• Assume for Chemelot
– Units A, B, C, D, E all 20 PJ and all 1.2 Mton
– Total = 100 PJ and 6 Mton (Chemelot footprint)
– Benchmark worldtop = 80 PJ = 4.8 Mton
• Present allocation Chemelot
– ßD + E (= 80/100 = 0.8) x 1.2 + ßC (=1.0) x 1.2 = 2.16 Mton
• Assume projects to reduce emissions
– Units A, B, C, D, E all 15 PJ and all 0.9 Mton
– Total = 75 PJ and 4.5 Mton (Chemelot footprint)
– Saving: 1.5 Mton
• Future allocation Chemelot
– ßD + E (= 80/75 = 1.07) x 0.9 + ßC (=1.0) x 0.9 = 1.86 Mton
– Allocation incentive: 0.3 Mton versus 1.5 Mton realised
26
Present ETS rules: new entrants & closures
 Unsolvable dilemmas new entrants (NE) & closures (C)
(see e.g. also Grubb and Neuhoff, Stern, Egenhofer, Weishaar, Matthes, Schyns,
Ecofys report for the EU Commission)
 Theory: freeze allocation [all allowances after C & zero for NE]
 Zero for NE actually hinders low carbon investments/competitiveness
 Retaining allowances after C – how long? – is worse than transfer
rules as it enhances market concentration
 Withdrawal allowances after C: perverse incentive keeping inefficient
plants in operation
 Most authors elaborate these problems, but fail to conclude
that within individual ex-ante frozen caps solutions are
simply impossible  search for squared circle
27
Basics of shortcomings present allocation
• Existing plants: allowances ex-ante frozen cap based on
historical emissions – rewarding pollution – frozen quantity,
whether production in- or decreases (“static, frozen economy”)
• New plants and debottleneckings: theory says buying (inhibits
efficient industry renewal); repair = allowances from a new
entrants’ reserve, also an ex-ante frozen cap (“plan-economy”)
• This principle = root cause of shortcomings, PLUS, as result:
– Insecurity investments in new plants (finite reserves)
– The allocation habit of few allowances for new plants versus many
allowances for existing plants : LACK OF EFFECTIVENESS to invest to
reduce emissions
– Repair: “transfer rules” (allowances closed plant to new efficient plant), but
new problem: high distortions, reinforcing market concentration
– Lowering production & selling freed allowances is declared equally
legitimate as investing to reduce emissions
28
Ex-ante rules prevent electricity liberalisation
• State interference prevents competitive market
– At gross margin of opportunity-cost, winning and losing market
share: zero sum game; at higher gross margin: distortions
– New entrants, vital for more competition, but ex-ante state decision
of operating hours determines profitability – plan economy
– Transfer rules protect incumbents: barrier to entry can be € 0.25
billion for a 1000 MWe power plant (4 years, or trading period)
– Even worse: incumbent does not apply for transfer rule and keeps
old plant stand-by (1000 MWe coal-fired plant of € 1.1 billion,
distortion ~ € 0.2 billion/year)
• Fight for allowances overrides fight for market share
• Price of system: economic rents – windfall profits
– Cause is the opportunity to sell allowances when not agreeing a
contract (opportunity-costs)
– Transfer of wealth to € 40-50 billion/year or double (EU-27)
29
Cap & trade: market price > opportunity-cost
Companies A & B: same production, efficiency
and same quantity of allowances
A
Euros
for an
equal total
production
volume
Companies
A & B
Mark-up
Gross
margin
cash
flow
Opportunity
cost
Cost of buying
allowances:
= distortion
Net profit
Net loss
B
Profit of
sales of
Allowances
= distortion
A wins
market
share
from B
30
Annexes
Benchmarking
Key principle of benchmarking
 What a CEO wants to know?
 He wants to know – e.g. with cost-price:
 Where his plants stand?; then
 Why? + What can be done about it?
 He refuses notions like “We are the best in the peer group of our [obsolete]
technology, or in our [small] scale, or in our plant vintage” (many
corrections make everyone equal)
 Key principle: benchmarks relate
 The product
… with
 the objective function – CO2 in the EU ETS
 Deviations shall be possible, but temporary and aimed to avoid
leakage outside EU (… objective function)
 Example: energy efficiency as objective function can avoid leakage by
switch to gas and shipping of carbon-rich fuels outside EU
32
Suitable benchmark formula
 Benchmark data of plants under the scheme (now EU)
Benchmark between average & best performance, e.g.
Benchmark = WAE – CF x (WAE – BP)
= WAE = Weighted Average Efficiency
= CF = Compliance Factor, to comply with total cap
= BP = proven Best Practice, proven means actual measured operational
data (or rather BP Group, for extra stimulation of innovation)
 Formula coincides with EU ETS Directive Annex III (3), average
emissions and achievable progress for each activity
 Industry opposes following alternative
 Related only to BP (BP + X%) – too short allocation, contra-incentive
to improve BP, effectiveness & innovation
33
Auctioning: clear incentive low carbon technologies, length trading
period irrelevant, but leakage & detrimental for competitiveness
High market liquidity
Buying allowances
Specific
energy use
or CO2
emission
Best
Practice
Weighted average
Incentive
Incentive
Decreasing efficiency order of plants
34
Performance-based trading: same incentive as auctioning, length
trading irrelevant, (hardly or) no leakage, good for competitiveness
High market liquidity
Buying allowances
Specific
energy use
or CO2
emission
Best
Practice
Free allocation
Selling allowances
Weighted average
Incentive
Incentive
benchmark
=
total
cap
Decreasing efficiency order of plants
35
Benchmark takes account of all energy carriers
Example steamcracker, simplified scheme
CO2
Feeds
(ethane,
LPG,
naphta,
gas oil,
etc.)
2/3 of the investment
Steam recovery
Furnaces with
heat recovery
to steam
Separations with
high power
compressors
Electricity
Methane from feedstock
Steam
Separation train can be:
• Efficient, with net-export
of steam of whole cracker
• Inefficient, steam import
• Both can be with the
same direct emission of
the cracker itself
Products
(ethylene, propylene, etc.)
Power train can be:
• Steam turbine driven
• Electric motor driven
• Combinations
High influence on
electricity & steam
balance, direct
emissions elsewhere
36
Benchmarks in a direct emissions scheme
 Easy inclusion in an ETS
 No conceptual problem in a direct scheme and no legal problem with
Directive, on the contrary
 Allowances according to deviation with benchmark
 In formula:
A = RDE + RSE – Σ production x (REE/RCE – benchmark) x CCF
= RDE = Realised Direct Emission (ton CO2)
= RSE = Realised Sequestered Emissions (ton CO2)
= REE/RCE = Realised Energy Efficiency (GJ/ton product) or Realised CO2
Efficiency (ton CO2/ton product)
= Benchmark = benchmark energy (or CO2) efficiency
= CCF = CO2 Conversion Factor (= 1.0 in case of CO2-benchmark)
 Note: Process emission is in this view included in the Best Practice
37
Misunderstandings power market cleared
• Fuel specific benchmarks: against objective function
= With ex-post: high fuel-switch prices, e.g. € 300-500/ton CO2
= Fuel switch limited with at least 50% (in case of 2 benchmarks)
= Coal plants without CCS encouraged (Carbon Capture & Storage)
• One electricity benchmark no deathblow coal-fired power
= Coal & lignite very important, climate policy means CCS !
= Cap & trade: opportunity-cost in power price (soft cost)
= With ex-post: CO2-cost in power price (real cost)
• Dash to gas with one benchmark?
=
=
=
=
Does not depend on one benchmark, but on total cap
Fuel switch at same CO2-price as cap & trade & auctioning
In fact more gas if more new coal and less CHP (given a total cap)
We need a controlled transition (CCS needs time)
38
References
•
•
References:
< http://www.dsm.com/en_US/html/sustainability/emission_trading.htm >
•
“Climate change challenges and the search for a sustainable policy”, 21 June
2005, 8th International Conference on Carbon Dioxide Utilisation (ICCDUVIII) 20-23 June 2005, Oslo, Norway.
“Options and consequences for the allocation of allowances to electricity
producers”, 21 December 2005, European Chemical Region Network (ECRN)
presidium meeting 21-22 December 2005, Maastricht, the Netherlands.
“Towards a simple, robust and predictable EU Emissions Trading Scheme –
Benchmarks from concept to practice”, 21 March 2006, presented to the
Dutch Ministry of Economic Affairs.
“The EU ETS is urgently in need of: effectiveness, level playing field,
competitiveness, fair & free competition”, 4th Congress of the ECRN, 10
November 2006, Tarragona, Spain, including:
– “One single benchmark for fossil-fuelled electricity in an Emissions
Trading Scheme: does it work, does it hurt and what about alternatives?”.
– “How to fit benchmarks with ex-post adjustments in the present EU
Emissions Trading Directive”.
•
•
•
39