C DMC
Retrofitting or Replacing:
The Decision of Installing Equipment
to reduce emissions from a coal-fired
power plant
Third Annual Carnegie Mellon Conference on the
Electricity Industry
March 14 2007
Dalia Patiño-Echeverri,
Benoit Morel, Jay Apt and
Chao Chen
1
C DMC
Motivation
The average U.S. coal-fired fleet is old and inefficient and many
plants are dirty and face the problem of
How to choose the best investment strategy given uncertainties on:
-Prices of emissions allowances for 3 P
-Future CO2 constraints
-Future Capital and OM for clean-coal technologies
-Fuel prices
2
C DMC
The Problem for the regulator
– How do current regulations affect industry decisions?
–
–
–
–
How do different factors affect capital investments decisions?
Interest rates?
Capital costs of new technologies?
Expectation of CO2 constraints? Timing and cost?
– Which policies can at the same time reduce emissions and
minimize cost of compliance in the future?
– Early announcement of mild carbon tax?
– Loan guarantees for the first n-of a kind IGCC?
– Subsidies for IGCC?
3
C
DMC
How should a “dirty” large-coal fired power
plant comply with current air emissions
regulations?
Assuming Cap and Trade and Coal
•
•
Reduce Emissions
–
Install Add-on controls
• WFGD to reduce SO2
• SCR to reduce NOx
• CI to reduce Hg
• CCS to reduce CO2 emissions
–
Replace the plant
• Super Critical Pulverized Coal (SC) (with or without CCS)
• Integrated Coal Gasification Combined Cycle (IGCC)
– With or without CCS
– CCS ready
Trade Emissions Allowances
4
C DMC
The Payoff of each strategy is uncertain
Capital
Costs
OM Costs
Risk of
shortages
and high
prices for
allowances
Fuel
Savings
Potential
affordable
CO2 capture
Trading
Allowances
Add-on
Emissions
Controls
Plant
replacement
5
C
DMC
A method to select the best investment
strategy (and understand effects of regulation)
Use Options and Forward Contracts Analogies
• Installing ECDs that can be turned on and off to
reduce emissions are equivalent to buying call
options for allowances
•
Inflexible ECDs provide the equivalent to forward
contracts for emissions allowances
6
C DMC
Options Analogy
1. Once installed the ECD can be operated or not
2. Operating the ECD has the same effect as buying allowances at the OMcost/ton.
Æ Installing the ECD = getting a bundle of call options on emissions allowances
T
Install ECD if capital cost ECD <
∫τ N Call (t , A , ω, X , r )dt
t
0
0
t
N t = Number of emissions reduced in year t
Call0 = Value at time 0 of a call option on an asset with :
current price A0
future price following a stochastic process ω
exercise price X t
risk - free rate r
7
C
DMC
Options and Forward Contracts Analogies to
quantify the benefits of the investment
•
Flexible ECDs: call options for allowances (WFGD, SCR)
•
Inflexible ECDs: forward contracts for emissions allowances (e.g. New
plants have less emissions because are more efficient.. But these
emissions reductions are not optional)
•
Multi-pollutant ECDs: basket options (e.g. WFGD reduces both SO2
and Hg emissions)
•
ECDs with requirements: Disjunctive Options (e.g. A CCS cannot be
operated without using the WFGD)
•
Investments that make possible other investments: Embedded Options
(e.g. The option to install CCS on a IGCC)
8
C DMC
Uncertainty on Allowances Prices
SO2 Allowances
NOx Allowances
Prices of SO2 Allowances. (Nominal dollars)
Prices of NOx Allowances. (Nominal dollars)
1800
5000
1600
Vintage 2004
Vintage 2005
Vintage 2006
Vintage 2007
4500
4000
1400
3500
1200
3000
1000
$/ton
$
$
$/ton
2500
2000
800
1500
600
1000
400
200
Jan04
500
Jan05
Jan06
Jan07
0
Jan04
Jan05
Jan06
Jan07
*Nominal Dollars
9
C DMC
Uncertainty on Allowances Prices
CO2 Allowances
Prices of CO2 Allowances. (British Pounds - Nominal)
30
25
£ /tonne
$
20
15
10
5
Apr05
Jul05
Oct05
Jan06
Apr06
Jul06
Oct06
Jan07
10
C
DMC
Characterizing the Uncertainty on Allowances
prices
Different Scenarios that specify the process followed by allowances
prices for SO2, NOx, Hg and CO2
– All price processes are assumed to be Geometric Brownian
σμ Motion (drift, volatility) with jumps
– Jumps account for changes in regulation that affect the new
starting price of allowances and the parameters of the GBM
95% Confidence Interval for future CO2 Allowances Prices for scenario # 2
35
$
30
95% Confidence
Interval for CO2
price
$ (2007)
25
20
15
Jump in 2025
10
5
0
2000
GBM low volatility
2010
2020
2030
year
Year
2040
2050
2060
11
Baseline analysis
C DMC
•
Plant:
– 1728 MW net output
– Appalachian-Medium Sulfur Coal
– Capacity Factor 65%. OM costs given by IECM
– 10,019Btu/kWh (30% of US plants are more efficient)
•
Capital and OM Costs for new plants as given by IECM
/ 30% higher for SC and IGCC
– Retrofit factors:
• WFGD and SCR on old plant Æ1.2
• CCS Æ 1.2 for old plant and SCPC
• CCS on IGCC Æ1.4 and 0 for IGCC cap ready
– OM Costs grow at inflation rate
– Capacity factor of 83% for SCPC and IGCC
– Coal 1.269$/MBtu
– Electricity $55/MWh
•
Allowances Prices of 3P
– Volatilities from historical Data
– Jump time from timetable of new regulations
– Jump size from AEO 2006 report
•
Scenarios for Prices for CO2
– No price
•
Discount Rate
– RFR + 3%
/ Different jump times, jump sizes and volatility
12
C DMC
Baseline Scenario Results
Investment Value (Net Benefits – Capital Cost)
x 10
No
CO2
. CO2-->
0 σ =0 Price
δ =0 t=0 0 σ =0 δ =0 t=0 0 σ =0 δ =0
9
W FGD
SCR
W FGD +SCR
W FGD+SCR+CCS
SCPC
SCPC+CCS
IGCC-ready
IGCC+CCS
IGCC
5
4
3
Value ($)
$ Benefits
Investment
CO2
jumps
to 44.8875
$21/tonne
in 2025
CO2-->
0 σ =0 δ =0 t=19
σ =0.05 δ =0 t=0 0 σ =0 δ =0
8
2
9
x 10 .
W FGD
SCR
W FGD +SCR
W FGD+SCR+CCS
SCPC
SCPC+CCS
IGCC-ready
IGCC+CCS
IGCC
6
4
Investment
$ Benefits
6
1
Value ($)
0
2
0
-1
-2
-2
-3
-4
0
5
10
15
20
25
T
30
35
40
Planning Horizon (Years)
45
50
-4
0
5
10
15
20
25
T
30
35
40
45
50
Planning Horizon (Years)
13
C DMC
Different scenarios for CO2
Volatility = 0.05
x 10
14
W FGD
SCR
W FGD +SCR
W FGD+SCR+CCS
SCPC
SCPC+CCS
IGCC-ready
IGCC+CCS
IGCC
5
4
8
$
6
1
4
0
2
1
2
3
4
5
Scenarios
6
7
8
9
2025 2020 2015 2010 2025 2020 2015 2010
$21/tonne
$58/tonne
Scenarios for CO2
T = T=40
Years
40 Several
Scenarios
9
W FGD
SCR
W FGD +SCR
W FGD+SCR+CCS
SCPC
SCPC+CCS
IGCC-ready
IGCC+CCS
IGCC
10
2
-1
x 10
12
$
3
$
TT==20 20
SeveralYears
Scenarios
9
$
6
0
1
2
3
4
5
6
7
8
9
2025 2020 2015Scenarios
2010 2025 2020 2015 2010
$21/tonne
$58/tonne
Scenarios for CO2
14
C DMC
Capital Cost of New Plants are 30% Higher
x 10
No
CO2
. CO2-->
0 σ =0 Price
δ =0 t=0 0 σ =0 δ =0 t=0 0 σ =0 δ =0
9
W FGD
SCR
W FGD +SCR
W FGD+SCR+CCS
SCPC
SCPC+CCS
IGCC-ready
IGCC+CCS
IGCC
5
4
3
$ Benefits
2
Investment
Value ($)
CO2
to44.8875
$21/tonne
in 2025
CO2--> jumps
0 σ =0 δ =0 t=19
σ =0.05 δ =0 t=0 0 σ =0 δ =0
8
1
9
x 10 .
W FGD
SCR
W FGD +SCR
W FGD+SCR+CCS
SCPC
SCPC+CCS
IGCC-ready
IGCC+CCS
IGCC
6
4
Investment
Value ($)
0
$ Benefits
6
2
0
-1
-2
-2
-3
-4
0
5
10
15
20
25
T
30
35
Planning Horizon (Years)
40
45
50
-4
0
5
10
15
20
25
T
30
35
40
45
50
Planning Horizon (Years)
15
C
DMC
Capital Costs New Plants are 30% Higher
CO2 volatility = 0.05
T=
20
Years
T= 20
Several
Scenarios
9
5
x 10
WFGD
SCR
WFGD +SCR
4
x 10
WFGD
SCR
WFGD +SCR
WFGD+SCR+CCS
SCPC
SCPC+CCS
IGCC-ready
IGCC+CCS
IGCC
12
WFGD+SCR+CCS
SCPC
SCPC+CCS
3
10
IGCC-ready
IGCC+CCS
$
1
$
8
$
$
IGCC
2
6
0
4
-1
2
-2
1
2
3
4
5
Scenarios
6
7
8
9
2025 2020 2015 2010 2025 2020 2015 2010
$21/tonne
T =T=40
Years
40 Several
Scenarios
9
14
$58/tonne
Scenarios for CO2
0
1
2
3
4
5
Scenarios
6
7
8
9
2025 2020 2015 2010 2025 2020 2015 2010
$21/tonne
$58/tonne
Scenarios for CO2
16
C DMC
CO2 volatility = 0.75
T = 20 Years
9
4
3
$
$
15
WFGD
SCR
WFGD +SCR
WFGD+SCR+CCS
SCPC
SCPC+CCS
IGCC-ready
IGCC+CCS
IGCC
5
T =T=40
Years
40 Several
Scenarios
9
T= 20 Several Scenarios
x 10
WFGD
SCR
WFGD +SCR
WFGD+SCR+CCS
SCPC
SCPC+CCS
IGCC-ready
IGCC+CCS
IGCC
10
$
2
$
6
x 10
5
1
0
-1
0
1
2
3
4
5
Scenarios
6
7
8
9
2025 2020 2015 2010 2025 2020 2015 2010
$21/tonne
$58/tonne
Scenarios for CO2
1
2
3
4
5
Scenarios
6
7
8
9
2025 2020 2015 2010 2025 2020 2015 2010
$21/tonne
$58/tonne
Scenarios for CO2
17
C DMC
Conclusions
•
For planning horizons longer than ~35 years, efficiency improvements
justify the replacement of old-plants (with efficiencies < than 70th
percentile) even in the absence of CO2 constraints
– Less efficient plants require shorter planning horizons
•
For short planning horizons, strong and early CO2 constraints are
required to justify plant replacement
•
Short planning horizons and other factors (risk-premium for cost of
new technologies, high interest rates) favor retrofits
•
Once old and inefficient plants are retrofitted with equipment to abate
SO2 and NOx it might be more difficult to make the case for the
reduction of CO2 emissions
18
C DMC
dpe@andrew.cmu.edu
19