Evaluating the Potential Impact of Transmission ... · e We d h u Fr i Satn Mon ue d u F r i Sat...
Transcript of Evaluating the Potential Impact of Transmission ... · e We d h u Fr i Satn Mon ue d u F r i Sat...
Evaluating the Potential Impact of Transmission Constraints on the Operation of a Competitive Electricity Market in Illinois
prepared for
Illinois Commerce Commission
June 2006
Richard R. Cirillo
Decision and Information Sciences Division
phone: 630-252-5629 email: [email protected]
The Study Was a Joint Effort:Argonne National LaboratoryUniversity of Illinois at Urbana-Champaign
Work commenced July 2002
Initial draft report submitted December 2003
Revised draft report with additional analysis September 2004
Briefing to Transmission Advisory Group May 2005
Final TAG comments received January 2006
Letter response to TAG comments sent March 2006
Final report with editorial changes April 2006
The Study Was Commissioned to Address Some Important Questions Regarding the Electricity Market in Illinois
Is the transmission system adequate to support market operation?
Can competition keep prices in check?
Is there the potential for market power to be exerted?– Market power: The ability to raise prices and increase profitability by
unilateral action
The Study Was Designed to Be:
– Exploratory, not predictive of specific outcomes
– Initial investigation
– Issue-oriented, not regulatory-oriented
Basic Assumptions Used in the Study
Analysis year : 2007
Projections for 2007 based on data from NERC, FERC, EIA, IEPA
A single day-ahead pool market operating in the State– All buyers and sellers of electricity bid into a common market every day
No bilateral contracts
Out-of-state generation and load participate in the market
Groups of cases– Case Study Assumptions – with generator forced outages– Conservative Assumptions – no forced outages
In the Pool Market, a Locational Marginal Price (LMP) Settlement Process Was Assumed to Be Operative
The cost of providing the next MW of power at each point in the network (i.e., the LMP) is calculated
In the absence of transmission limitations:– The lowest cost generators are used first– Higher cost generators are used only as
needed– The LMP at every point in the network is
the same
With transmission limitations:– It is not always possible to use the lowest
cost generators– The LMPs vary, sometimes considerably,
across the networkTransmission
Nodes
Cases Were Constructed as “Electronic Experiments”to Study Market Behavior
Production Cost •Generation companies bid production cost•Gives the lowest system cost
Physical Withholding •Intentionally take generators out of service•Prices rise
Economic Withholding •Generation companies bid above production cost•Prices rise
•Experiments moved from very simple to more complex strategies
•Production Cost Case used as a benchmark
•Not intended to imply that any company would attempt to exercise market power
•Only an initial mapping of possible market bidding
In the Production Cost Case for the Analysis Year 2007, the State Is a Net Exporter of Electricity
0
2,000
4,000
6,000
8,000
10,000
12,000
14,000
16,000
18,000
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
In-State Generation
(GWh)Net ExportIn-State Demand
Approximately 6% of the electricity generated is exported
Coal and Nuclear Capture the Largest Market Share
0
2
4
6
8
10
12
14
16
18
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
Generation(TWh)
Renewable
Oil
Nat Gas
Coal
Nuclear
Approximately 2% of the electricity is from natural gas and other sources
In the Production Cost Case, 4 out of 22 Companies Account for About 95% of the Market
Out-of-state suppliers do not gain significant market share
Generation Company
Market Share Exelon Nuclear43.1%
Ameren20.1%
Midwest Generation
16.8%
DynegyMidwest
Generation14%
Dominion Energy3.4%
Other 1.5%
City of Springfield1.0%
0
200
400
600
800
1,000
1,200
1,400
1,600
Exelon N
uclear
Ameren
Midwest G
enera
tion LLC
Dyneg
y Midwes
t Gen
eratio
n Inc.
Dominion Energy
Exelon N
uclear/
Midameri
can Energ
y
City of S
pringfie
ld
NRG Energy
Relian
t Energ
yCalp
ine
Duke Energ
y
Dyneg
y/NRG Energ
y
Constella
tion Power
Southern Ill
inois Power
Coop
MidAmerica
n Energy C
o
Alleghen
y Power
Aquila Energ
yPPL
Power Energ
y Part
ners
Soyland Power
Coop Inc
Calumet
Energy L
LC
Southwester
n Electri
c Coop.
Revenuesand
Costs($ million)
RevenuesCosts
Only 5 of 22 Generation Companies Show an Operating Profit In the Production Cost Case. Others Are Not Dispatched Enough to Recover Fixed Operating Costs.
LMPs at Transmission Nodes Increase In the Peak Load Summer Months and Vary Across the State, Even Under Production Cost Conditions
January February March April JuneMay
July August September October November December
HigherLowerLMP Price and Duration:
Four Downstate Zones in April (Low Load Month)LMPs Are Low and Close Together
0
20
40
60
80
100
120
140
160
180
200
SunMonTueWedThu Fri SatSunMonTueWedThu Fri SatSunMonTueWedThu Fri SatSunMonTueWedThu Fri SatSunMon
LMP($/MWh)
Little Transmission Congestion
Four Downstate Zones in July (High Load Month)LMPs Are a Little Higher But Still Close Together
0
20
40
60
80
100
120
140
160
180
200
SunMonTueWedThu Fri SatSunMonTueWedThu Fri SatSunMonTueWedThu Fri SatSunMonTueWedThu Fri SatSunMon
LMP($/MWh)
Little Transmission Congestion
Six Northern Zones in April (Low Load Month)LMPs Are a Little Higher than Downstate But Close Together
0
20
40
60
80
100
120
140
160
180
200
SunMonTueWedThu Fri SatSunMonTueWedThu Fri SatSunMonTueWedThu Fri SatSunMonTueWedThu Fri SatSunMonTue
LMP($/MWh)
Little Transmission Congestion
Six Northern Zones in July (High Load Month)LMPs Are High and with a Significant Spread
0
20
40
60
80
100
120
140
160
180
200
SunMonTueWedThu Fri SatSunMonTueWedThu Fri SatSunMonTueWedThu Fri SatSunMonTueWedThu Fri SatSunMonTue
LMP($/MWh)
Noticeable Transmission Congestion
Physical Withholding Cases Were Studied
Withhold generator units from the market
– Reduce supply to cause prices to rise
– Increase profitability of other units owned by the company
Cases studied:
– Withhold one unit at a time: Does not increase profitability
– Use a profitability screen: Does not significantly increase profitability
– System reserve screen: Can increase profitability
Units Withheld Using the System Reserve Screen Could Potentially Increase Prices and Profitability
•Impact of physical withholding depends on:– Capacity withheld– Location on the network– Availability of replacement capacity
•Increase in company peak day operating profit: $2-14 million
•Increase in peak day consumer costs: $45-230 million
•Conservative assumptions do not change the basic result
•This assumes there is no market oversight or prohibition on withholding•There are some physical constraints on the ability to withhold large units•There is no implication that any company would do this
Economic Withholding Cases Were Studied
Generator unit prices are increased– Units are not taken out of service
– Prices increased relative to production cost
Cases studied– Single unit price increases: Does not increase profitability
– Companywide price increases in all hours: Profitability increase but loss of market share
– Companywide price increases in peak hours: Profitability increase with little loss of market share
For Some Companies, Increasing Prices All Day Leads to a Loss of Generation Market Share But Can Increase Profitability
0
2
4
6
8
10
12
14
16
18
20
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
Companywide Price Increase (Multiple of Production Cost)
Peak Day Operating
Profit ($million)
0
50
100
150
200
250Peak Day
Generation (GWh)
Profit
Generation
All Day Price Increases as a Multiple of Production Cost
Peak Day Generation(GWh)
Peak Day Operating Profit($million)
Company rate of generation drops as prices are increased, then levels off when units are needed to meet peak load
Company profitability increases only slightly until units are needed to meet load, then increase significantly
For Some Companies, Increasing Prices Only During Peak Hours Minimizes the Loss of Market Share
0
2
4
6
8
10
12
14
16
18
20
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
Companywide Price Increase (Multiple of Production Cost)
Peak Day Operating
Profit ($million)
0
50
100
150
200
250Peak Day
Generation (GWh)
Profit
Generation
Peak Hour Price Increases as a Multiple of Production Cost
Peak Day Generation(GWh)
Peak Day Operating Profit($million)
Company rate of generation drops only slightly as prices are increased
Company profitability increases steadily
Overall Observations Under the Assumed Conditions
The State has an abundance of generation capacity. The State is a net exporter of electricity. Virtually all the generation is supplied by coal and nuclear; only a small amount by natural gas.
Using several indices, the generation market in the State is highly concentrated.
The transmission system is adequate for most hours but will show signs of congestion about 5% of the time. During high load periods, transmission congestion causes prices to rise at different rates across the State, even with no exerting of market power.
There is the potential for some companies to exercise market power (i.e., raise prices and increase profitability by unilateral action) and raise consumer costs under selected conditions, particularly when there is transmission congestion.
The Conservative Assumptions (no forced outages) do not change the basic results.
Additional Work Is Needed to Expand the Understanding of the Electricity Market
Improved representation of out-of-state generation and load
Sensitivity analyses
Effect of bilateral contracts
Effect of consumer price responsiveness
Addition of generation and transmission resources in key places
Effect of market rules
Comments Received on the Study Made Several Major Points
The study does not reflect the current operating practices of the PJM/MISO markets– Agree – Modeling of alternative market rules was originally proposed as a
later phase of the study.
The data and information used in the report have been superseded– Agree– The study was not intended to be a one-time effort. The plan was to
have ICC staff use the model and data in-house for updating and further analysis.
The results have no relevance to the current situation– Strongly disagree
Examples of Generator Bids in PJM for July 26, 2005Illustrate Bidding That Goes Beyond Production Cost
8X2D
0
200
400
600
800
1000
0 10 20 30 40 50 60Amount (MW)
Pric
e ($
/MW
)
2O
0
200
400
600
800
1,000
1,200
0 2,000 4,000 6,000 8,000 10,000Amount (MW)
Pric
e ($
/MW
)
On a companywide basis, GenCo 2O bid almost 9,000 MW into the market. Approximately the last 1,000 MW of the company’s capacity were bid at prices greater than 600 $/MW.
GenCo 9O bid unit 8X2D of 48 MW at 800-900 $/MWh
•These are actual bids, not model results•It is not possible to determine if the bids were accepted•Market monitor actions in response to the bids are unknown
PJM Data on LMPs for July 26, 2005 Show the Effects of Generator Bids and Transmission Congestion
Example Individual Bus LMPs
0
100
200
300
400
500
600
700
800
900
1000
0 200 400 600 800 1000 1200 1400 1600 1800 2000 2200 2400
Hour of the Day
LMP($/MW)
BIXBYMOUNDHOCKINGHOCKINGMONTGOMEHILLVALLBAKER
Bus
ComEd Bus and Zone LMPs
0
20
40
60
80
100
120
140
160
180
0 200 400 600 800 1000 1200 1400 1600 1800 2000 2200 2400Hour of the Day
LMP($/MW)
Lowest Bus - 176 STILHigest Bus - 389 E ROComEd Zone
Selected ComEd LMPsSelected PJM LMPs
There Is Strong Agreement with Several Comments
“The most important finding of the study is the rigorous demonstration that the transmission system is a vital part of an energy market.”
- Ameren Electric
“…the necessity for RTOs to continue to refine the independence of their transmission system operations, market designs, market rules andoversight of the RTO market by market monitors.”
- PJM