The impact of carbon pricing on
industrial competitiveness
EPRG Spring Seminar, Cambridge
28th May 2007
Jean-Charles Hourcade, Damien Demailly,
CIRED
Michael Grubb, Misato Sato and Karsten Neuhoff
EPRG
Project convened by:
Contents
1. Introduction: basic analytics
2. Which (sub) sectors are affected?
3. Can we model market share impacts?
4. Better understanding the process chain..
Part 1
Basic analytics
In turn there are 3 key determinants of competitiveness
exposure
-
Higher
Low
High Lower
Value at stake as a
proportion of
sector profit
Ability to pass on costs to customers
Unaffected
Offset / gains? At risk?
Marginal impact
•Ferrous metals•Cement?•Oil refining•Glass•Aluminium•Chemicals
•Electricity•Cement?
•Paper (newsprint)
•Pharmaceuticals•Food & drink•Retail•Transport
•Pulp & Paper
-
Higher
Low
High Lower
Value at stake as a
proportion of
sector profit
Ability to pass on costs to customers
Unaffected
Offset / gains? At risk?
Marginal impact
•Ferrous metals•Cement?•Oil refining•Glass•Aluminium•Chemicals
•Electricity•Cement?
•Paper (newsprint)
•Pharmaceuticals•Pharmaceuticals•Food & drink•Food & drink•Retail•Retail•Transport•Transport
•Pulp & Paper
1. Energy intensity of
production;
2. Ability to pass
through cost
increases to prices;
3. Abatement
opportunities
A sector’s cost is potentially impacted by ETS via 3 channels:
1. Cost of the emissions
2. Costs of abatement
3. Increased electricity prices
Part 2
Who is affected?
Upper end of range: zero free allocation
Lower end of range: 100% free allocation
Assumptions: CO2 price=€15/tCO2; Pass through in electricity = €10/mwh
Using 4 digit (SIC 92) representation of the sectors…
Cement
Electriciy
Textiles
Food & Tobacco
Pulp &
paper
Refining & fuels
Chemicals &
plasticsGlass &
ceramics
Iron & Steel
Non-ferrous metals
including Aluminium
Metal
manufactures
0%
5%
10%
15%
20%
25%
0 10 20 30 40
UK trade intensity from outside the EU
Potential value at stake (NVAS/MVAS) under 0
to 100% free allocation
MVAS: Max. value at stake
(no free allocation)
NVAS: Net value at stake
(100% free allocation; exposure to
electricity price only)
Vertical range gives
insights on:
- Marginal cost
increase (top end
of bar)
-Impact of electricity
price pass through
Combined with
horizontal axis gives
insights on:
-Scope of
auctioning
- dynamics of
impacts
Upper end of range: zero free allocation ; Lower end of range: 100% free allocation
Assumptions: CO2 price=€15/tCO2; Pass through in electricity = €10/mwh
Setting against the trade intensity from other EU countries
gives insight into the potential degree of concern about
differential allocation between Member States.
Cement
Electriciy
Textiles
Food & tobacco
Pulp &
paper
Refining & fuels
Chemicals &
plastics
Glass &
ceramics
Iron & Steel
Non-ferrous metals
including Aluminium
Metal manufactures
0%
5%
10%
15%
20%
25%
0 5 10 15 20 25 30 35 40 45
UK trade intensity from within the EU
Potential value at stake (NVAS/MVAS) under 0 to
100% free allocation
MVAS: Max. value at stake
(no free allocation)
NVAS: Net value at stake
(100% free allocation; exposure
to electricity price only)
Moving from a 3 digit to a 4 digit (SIC 92) representation of
the sector e.g. break-down of Iron & Steel sector (non-EU trade)
0%
2%
4%
6%
8%
10%
12%
14%
16%
18%
20%
0% 10% 20% 30% 40% 50%
Derived Non-EU Trade Intensity
Potential value at stake (NVAS / MVAS) under 0 to 100% free allocation
-Man. of basic iron & steel & of ferro
(alloys (ECSC
Man. of cast iron tubes
Man. of steel tubes
Cold drawing
Cold rolling of narrow strip
Cold forming or folding
Wire drawing
1064
82
7069 359
22
47
MVAS: Max. value at stake
(no free allocation)
NVAS: Net value at stake
(100% free allocation;
exposure to electricity price
only)
GVA (£million, 2004)
0%
5%
10%
15%
20%
25%
30%
0% 20% 40% 60% 80%Derived Non-EU Trade Intensity
Potential value at stake (NVAS / MVAS) under 0 to 100% free allocation
Precious metals production
Aluminium production
Lead, zinc and tin production
Copper production
Other non-ferrous metal
production
135
444
131
65988
MVAS: Max. value at stake
(no free allocation)
NVAS: Net value at stake
(100% free allocation;
exposure to electricity price
only)
GVA £million, 2004
0%
1%
2%
3%
4%
5%
6%
7%
8%
9%
10%
0% 20% 40% 60%Derived Non-EU trade
Intensity
Potential value at stake (NVAS / MVAS) under 0 to 100% free allocation
Manu. of industrial gases
Manu. of dyes and pigments
Manu. ofother inorganic basic chemicals
Manu. of other organic basic chemicals
Manu. of fertilizers and nitrogen compounds
Manu. of synthetic rubber in primary forms
Manu. of pesticides and other agro-chemical
productsManu. of paints, varnishes and similar
coatings, printing ink and masticsManu. of basic pharmaceutical products
Manu. of pharmaceutical preparations
Manu. of perfumes and toilet preparations
Manu. of explosives
Manu. of glues and gelatines
Manu. of soap and detergents, cleaning and
polishing preparationsManu. of essential oils
Manu. of photographic chemical material
Manu. of prepared unrecorded media
.Manu. of other chemical products n.e.c
Manu. of man-made fibres
Manu. of rubber tyres and tubes
Retreading and rebuilding of rubber tyres
Manu. of other rubber products
Manu. of plastic plates, sheets, tubes and
prof ilesManu. of plastic packing goods
Manu. of builders' w are of plastic
Manu. of other plastic products
Manu. of plastics in primary forms
393
450
349
29
156
562
212
351
1159
1011
4350
5938
1860
1708
1990
2105
1367
GVA £million, 2004
Non-ferrous metals Chemicals & plastic
0%
5%
10%
15%
20%
25%
0% 20% 40%
Derived Non-EU Trade Intensity
Potential value at stake (NVAS / MVAS) under 0 to 100% free
allocation
Manu. of flat glass
Shaping and processing of flat
glass
Manu. of hollow glass
Manu. of glass fibres
Manufacture and processing of
other glass, including technical
glasswareManu. of ceramic household and
ornamental articles
Manu. of ceramic sanitary fixtures
Manu. of other ceramic products
Manu. of refractory ceramic
products
Manu. of other technical ceramic
products
Manu. of ceramic tiles and flags
329
15988
527
157
227
288159
178
22
GVA £million, 2004
17 (1999 data)
Glass & ceramics
0%
1%
2%
3%
4%
5%
6%
7%
8%
9%
10%
11%
12%
13%
14%
15%
0% 10% 20%
Derived non-EU Trade Intensity
Potential value at stake (NVAS / M
VAS) under 0 to 100% free allocation
Manu. of bricks, tiles and
construction products, in baked
clayManu. of cement
Manu. of lime
Manu. of plaster
Manu. of concrete products for
construction purposes
Manu. of plaster products for
construction purposes
Manu. of ready-mixed concrete
Manu. of fibre cement
,Manu. of other articles of concrete
plaster and cement
Cutting, shaping and finishing of
stone
Production of abrasive products
Manu. of other non-metallic mineral
.products n.e.c
409
9
403
16
103268
272
34
77
375556
1195
GVA £ million, 200466.7%
Cement
Conclusion 1.
The analysis at 4-digit level identifies 2 groups of potentially
exposed sub-sectors, with some overlap:
A) Indirectly exposed (electricity intensive production):
Top 5: production of precious metals; manufacture of
industrial gases; other inorganic basic chemicals; other
technical ceramic products; household and sanitary goods;
B) Potentially directly exposed (carbon intensive production)
Top 5: manufacturing of lime; production of precious metals;
other technical ceramic products; basic iron & steel;
manufacturing of cement.
�significant impact of ETS on competitiveness concentrates
on a far smaller fraction of industrial activities than
suggested by aggregate figures.
Conclusion 2.
• Overall, 20 out of 92 sub-sectors fall under either/both:
A)>1.5% electricity impacts at €15/t CO2;
B)> 3% Maximum potential NVAS
(i.e.CO2 price of €50/t CO2, would therefore correspond to
exposure of 5% and 10% respectively. )
• For the UK, the combined Gross Value Added of the top 20
potentially exposed is small (around 1% of total UK GVA).
• low overall impact on total GVA of economy �implies low
political obstacles towards finding international solutions to
address competitiveness concerns for these sectors
Part 3
Can we model EU ETS impacts on
market share?
How will climate policies impact the RIP?
Cement∆Production cost
betweenEU and RoW
Transport
Consumption / Capacity
Import restriction
Cost of Instability
Product differentiation
Service differentiation
International pressure on EU
market
+
+
+-
Trade Barriers
No ETSETS
Climate Policies
International pressure on EU
marketEU cost increase
due to ETS?
?
?
How will climate policies impact the RIP?
Steel
+
+
+
-
ETS
Climate Policies
International pressure on EU
market
International pressure on EU
market
∆Production cost
betweenEU and RoW
Transport
Consumption / Capacity
Import restriction
Cost of Instability
Product differentiation
Service differentiation
Trade Barriers
No ETS
-
+
EU cost increase
due to ETS
?
Central scenario
EU Import ratio
0
5
10
15
BAU 15 30 45
CO2 price (€/tCO2)
%
Steel Cement
Central scenario
Gross Profit margin for various Rates of Free Allocation (RFA)
0
15
30
BAU 15 30 45
CO2 price (€/tCO2)
%
100%
95%
90%
75%
50%
0%
-15
0
15
30
45
60
75
BAU 15 30 45
CO2 price (€/tCO2)
%
100%
95%
90%
75%
50%
0%
STEEL
CEMENT
Central scenario
40
45
50
55
60
65
15 30 45
CO2 price (€/tCO2)
%
Steel Cement
Compensating Rate of Free Allocation
Conclusions
According to the econometric estimates of key parameters, and given their
uncertainty:
• Market share losses are likely to remain modest
• Market share losses in the cement sector are of the same order of
magnitude than in the steel sector, the high CO2-intensity of the former
offsetting its lower trade sensitivity
• The CRFA : it is all about the rate of cost pass through (PT)… Hence,
huge uncertainty.
• A “wrong” CRFA has drastic impact on the cement sector’s profitability,
much less for the steel sector
• Finally, what RFA? It is all about your risk aversion…
Part 4
Towards better understanding the
Production Chain:
Basic Oxygen Furnace production
0%
10%
20%
30%
40%
50%
60%
0 500 1000 1500 2000
GVA (million £)
Electricity VAS/ Co2 VAS
NVA electricity
NVA totalslab
basic iron&
steel
iron&
steel
NVA electricity (from slab)
NVA total (from slab)
Note – preliminary results – particularly GVA slab pure guess.
This has implications for VAS estimations for slab.
Cement
NVA electricity
NVA total
0%
5%
10%
15%
20%
25%
30%
0 500 1000 1500 2000
GVA (million £)
Electricity VAS/ CO2 VAS
NVA electricity (from clinker)
NVA total (from clinker)
Clinker
Cement
Concrete products (concrete
products for construction; mixed
Note – preliminary results – particularly electricity split clinker/cement pure guess
Part 5
Annex
4 digit analysis : new approach to defining trade intensity
For the 3 digit analysis, we define UK trade intensity from the EU as:
= value of imports from EU + value of Exports to EU
value of total UK market value
For market value we use total supply=total demand from Input Output tables
Due to data constraints at 4 digit level, in this analysis we use:
= Value derived EU exports + Value derived EU imports
annual turnover + val. total imports - val. total exports
Where we define:
Value derived EU exports = Total exports at 4 digit × Exports to EU 3 digit
Total export 3 digit
Long products Flat products
• Low value added products and
differentiation (?)
• High transportation cost for scrap
steel
�Local market: EU Import ratio ~ 10%
• High product differentiation
• Three Regional markets (Asia, North
America and Europe) partially linked
• EU Import ratio remains modest
(~10%)
• Price differences maintain
0
20
40
60
80
100
120
140
mai-05 sept-05 déc-05 mars-06 juil-06 oct -06 janv-07
FLAT PRODUCTS WORLD
FLAT PRODUCTS EU
FLAT PRODUCTS N. AMERICA
FLAT PRODUCTS ASIA
International pressure on
the EU Steel sector
Is this situation sustainable?
A possible new scheme: slab production in low cost countries, product
differentiation close to consumers
Modelling assumptions
• Time Horizon: 2015
• Geographical aggregation : EU 27
• Products: flat and long steel products are aggregated
• For a given CO2 price, 3 elements in the cost increase due to the ETS:
• Electricty cost increase (full pass-through in the electricity sector)
• Abatement cost: depends on the Marginal Abatement Cost Curve (MACC)
• Emission cost: free allowances (if any) are purely grandfathered
• Price increase: depends on the Pass Through (PT)
• Market share loss: depends on the trade elasticity (σ)
• Demand drop: depends on the demand elasticity (ε)
KEY
PARAMETERS
Central scenario
Values for key parameters = the range mean
0
10
20
30
40
15 30 45
CO2 price (€/tCO2)
Cost increase
(€/t of cement)
∆ec uac PCO2*ue
20
40
60
80 Total costincrease(%)
0
∆c/c0
0
10
20
30
40
15 30 45
CO2 price (€/tCO2)
Cost increase
(€/t of cement)
∆ec uac PCO2*ue
20
40
60
80 Total costincrease(%)
0
∆c/c0∆c/c0
0
10
20
30
40
15 30 45
CO price (€/tCO2)
Cost increase
(€/t of steel)
∆ec uac PCO2*ue
5
10 Total costincrease(%)
0
∆c/c0
0
10
20
30
40
15 30 45
CO price (€/tCO2)
Cost increase
(€/t of steel)
∆ec uac PCO2*ue
5
10 Total costincrease(%)
0
∆c/c0∆c/c0
Cement Steel
Cost Impact
CAVEATS
Caveat N°°°°1 : Uncertainty surrounding these parameters� For every parameter, we test a range of values (from economics literature)
and define a density of probability
1.5
2.8
full
SteelCement
PRIMES +/- 33%
0.2
0.5
0
122Trade
elasticity
10Demand
elasticity
0
MACC
fullPT
0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.00.000
0.001
0.002
0.003
0.004
0.005
0.006
0.007
Uncertainty on the trade elasticity
Steel Cement
50
60
70
80
90
100
EU DomesticMarketshare(%)
50
60
70
80
90
100
EU DomesticMarketshare(%)
100 104 108 112 116 120
σ=12
σ=7
σ=2
100 104 108 112 116 120
σ=12
σ=7
σ=2
EU Price / Initial EU Price
78
80
82
84
86
88
90
92
94
96
98
EU DomesticMarketshare
(%)
78
80
82
84
86
88
90
92
94
96
98
EU DomesticMarketshare
(%)
100 110 120 130 140 150 160
EU Price / Initial EU price * 100
σ=2.8
σ=1.75
σ=0.7
100 110 120 130 140 150 160
EU Price / Initial EU price * 100
σ=2.8
σ=1.75
σ=0.7
EU Price / Initial EU Price
Cement
Uncertainty on the pass through
A controversial issue…
Theoretical Literature enhances the paradoxical role of market power
� Trade exposure is not the only PT determinant
Empirical literature:
• Ex-ante studies use a wide range of estimates
• Econometric works claim for significant PT:
• Walker: PT of the CO2 opportunity cost in 2005 from 10 to 40% in the
cement sector
• Literature on exchange rate � PT on export markets from 20 to 70% for
these two sectors
CAVEATS
Problem N°°°°2 : reliability of econometric estimates
• Poor estimates availability for the EU
• Estimates based on small shocks
• Estimates based on past data, whereas the determinants of trade evolve (e.g. slab
trade)
• No distinction between trade barriers (all mixed) whereas they will evolve differently
over time
• Do not take into account the impact of climate policies on trade barriers
Nevertheless…
Sensitivity Analysis
∆ Import ratio0% 1% 2% 3% 4%
PT
σ
MACC
ε
∆ Import ratio0% 1% 2% 3% 4%
PT
σ
MACC
ε
CRFA25% 50% 75% 100%
PT
ε
σ
MACC
0% CRFA25% 50% 75% 100%
PT
ε
σ
MACC
0%
Range
Likely range
Mean
Range
Likely range
Mean
CEMENT
0
0,1
0,2
0,3
0,4
0,5
0,6
0,7
0,8
0,9
1
0 2 4 6 8 10 12 14
Delta Import ratio
PCO2=15
PCO2=30
PCO2=45
PCO2
0
0,1
0,2
0,3
0,4
0,5
0,6
0,7
0,8
0,9
1
0 2 4 6 8 10 12 14 16
Delta Import ratio
PCO2=15
PCO2=30
PCO2=45
PCO2
Multi-sensitivity Analysis
STEEL
Mean: -2%
CEMENT
Mean: -2%
∆ EU Import ratio
Repartition function
0
0,1
0,2
0,3
0,4
0,5
0,6
0,7
0,8
0,9
1
0 20 40 60 80 100 120
RFA (%)
∆µ=-15
∆µ=-5
∆µ=+5
∆µ=+15
CRFA
PCO2=45/30/15
∆µ=-10
∆µ=-20
∆µ=+10
∆µ=+20
0
0,1
0,2
0,3
0,4
0,5
0,6
0,7
0,8
0,9
1
0 20 40 60 80 100 120
RFA (%)
∆µ=-15
∆µ=-5
∆µ=+5
∆µ=+15
CRFA
PCO2=45/30/15
∆µ=-10
∆µ=-20
∆µ=+10
∆µ=+20
Multi-sensitivity Analysis
STEEL
Mean CRFA: ~60%
CEMENT
Mean CRFA: ~50%
CRFA
Repartition function
0
0,1
0,2
0,3
0,4
0,5
0,6
0,7
0,8
0,9
1
0 10 20 30 40 50 60 70 80 90 100 110 120
RFA (%)
∆µ=+5
CRFA
PCO2=45/30/15
∆µ=-5
0
0,1
0,2
0,3
0,4
0,5
0,6
0,7
0,8
0,9
1
0 10 20 30 40 50 60 70 80 90 100 110 120
RFA (%)
∆µ=+5
CRFA
PCO2=45/30/15
∆µ=-5
+
_
_
+
Mio. €
The current approach of free allocation
shields profits, not the production of effected
sectors
• energy intensive industry has usually high fixed costs
• relocating production is a strategic (long-term) decision
• competitiveness is affected by post 2012 perspective
0
2000
4000
6000
8000
10000
12000
10 20 30 40 50 €/tCO2
00 10 20 30 40 50
Allocation of
90% of historic
emissions
Revenue
minus operational costs
100% auction
Used allowances
allowances sold as production
is decreased
Example: Cournot model of the European cement sector
Assumptions: For 20€/tCO2, extended cost: +14€/t cement ~200km by road`
Phase I
2005-07
Phase II
2008-12
Continued international cost differences
effect energy intensive industry.
Global or sectoral
agreements
revenues finance investment
Robust solutions for post 2012 exist
Efficient production
Environmental costs
reflected in price
Fair competition
Compensation of
Exports/imports
Allocation pro-
portional to output
We will find the best solution in an international dialogue.
