dynamic pricing, tariffs, and price responsive demand programs real time pricing at georgia power...
Post on 21-Dec-2015
215 views
TRANSCRIPT
Dynamic Pricing, Tariffs, and Price Responsive Demand Programs
Real Time Pricing at Georgia Power Company and Duke Power Company
Michael T. O’SheasyChristensen Associates
September 9, 2002
September 2002 2 CHRISTENSEN ASSOCIATES
Topics
What is RTP and What are the Benefits? RTP at Georgia Power Company and Duke
Power Company Why does an RTP Product make sense (and
cents)? Features of a Two-Part RTP and resultant
prices Price Response and Market Effects PPP and Portfolio Pricing
September 2002 3 CHRISTENSEN ASSOCIATES
An electricity rate structure in which retail energy prices:
• Vary frequently (e.G., Hourly),
• With short notice (e.G., Hour-ahead or day-ahead),
• To reflect expected hourly costs
• It can mimic an open market and market prices
What is Real-Time Pricing?
September 2002 4 CHRISTENSEN ASSOCIATES
Benefits from a Real-Time Pricing Program
Improved system reliability Reduced wholesale price volatility Less reliance on outside (high-priced) power
purchases Typical utility will achieve, at least cost, a “virtual
generation” asset to achieve about 5% of total system load requirements in critical hours (about 20% load response of load participants total load)
September 2002 5 CHRISTENSEN ASSOCIATES
Benefits from a Real-Time Pricing Program
Customer satisfaction— Provides open access to market— Two-part structure limits price risk exposure
Facilitates economic growth— Industrial/commercial customers on RTP grow faster— Growth tends to occur off-peak— Utilities “earnings before income taxes” may increase
RTP customers have incentive to innovate with economic energy efficiency programs/devices
September 2002 6 CHRISTENSEN ASSOCIATES
The Duke Program
Program Began Fall of 1993Program Began Fall of 1993 Currently 53 CustomersCurrently 53 Customers Average PricesAverage Prices
—Transmission $.034Transmission $.034—Distribution $0.04Distribution $0.04
Demand Reduction 200 mW at High PricesDemand Reduction 200 mW at High Prices Two-Part Tariff DesignTwo-Part Tariff Design
Program Began Fall of 1993Program Began Fall of 1993 Currently 53 CustomersCurrently 53 Customers Average PricesAverage Prices
—Transmission $.034Transmission $.034—Distribution $0.04Distribution $0.04
Demand Reduction 200 mW at High PricesDemand Reduction 200 mW at High Prices Two-Part Tariff DesignTwo-Part Tariff Design
September 2002 7 CHRISTENSEN ASSOCIATES
Price Responsive Customers
Universities with Generators Textile and Chemical Fiber Plants with Generators Large Paper Manufacturer with Grinders Steel Mill and Electrode Manufacturer Each with
an Arc Furnace Brewery with Generator Food Processor with Generator 25 Customers Identified as Price Responsive
September 2002 8 CHRISTENSEN ASSOCIATES
Typical Duke Prices on High Price and Average Price Day
0
50
100
150
200
250
300
1 3 5 7 9 11 13 15 17 19 21 23 25
Hours
$/m
Wh
September 2002 9 CHRISTENSEN ASSOCIATES
RTP in the State of Georgia Year 2001
Largest Program in the World > 1600 Customers
— > 5,000 MW— > $1 billion revenue
IRP Resource Price Protection Products
September 2002 10 CHRISTENSEN ASSOCIATES
Real-Time Pricing for GPC
Two-part tariff design Day-ahead RTP
— 250 kW minimum Hour-ahead RTP
— 5 MW minimum
September 2002 11 CHRISTENSEN ASSOCIATES
GPC Philosophy on RTP
1. RTP is the marginal cost of providing electricity
• Lambda
• Losses
• Marginal Cost of Transmission
• Outage/ct Cost
• Risk Adder
September 2002 12 CHRISTENSEN ASSOCIATES
2. Powerful Product in a supplier’s Pricing Portfolio
• Customers Satisfaction
• Economic Development
• Customer Choice
• Constant Profit Contribution per kWh
• Credits below CBL
GPC Philosophy on RTP
September 2002 13 CHRISTENSEN ASSOCIATES
Why is There a Need for RTP?
1. Product characteristics create risk2. Many customers have on-site
flexibility which can manage risk Expected
Ave ¢/KWH
Hours
Target Customers Load Shape
KW
Off Peak
Off Peak
On Peak
Target Customers Actual Load Shape
Target Customers Forecasted Load Shape
On peak and off peak prices* based upon forecasted load shape
On peak and off peak prices* based upon actual load shape
Hours
¢/KWH
TOU Prices
Scenarios
*Assume Actual Hourly Prices equal Forecasted Hourly Prices
September 2002 14 CHRISTENSEN ASSOCIATES
Risk on Seller
100%, 100%
100%
100%0
•HUD
•CED
•TOU
•1 Part RTP
•2 Part RTP
•Curtailable EnergyLoad
Shape Risk on Seller
Cost Risk on Seller
•Flat Bill
•Flat Energy
September 2002 15 CHRISTENSEN ASSOCIATES
September 2002 16 CHRISTENSEN ASSOCIATES
Features of “Two-Part” RTP
Customer pays for a baseline level of usage (e.g., recent historical usage) at standard tariff prices. Customer is revenue neutral at baseline usage.
Differences in usage from the baseline (increases or decreases) are billed at RTP prices.
Demand response benefits the RTP customer, the utility, and all other customers.
September 2002 17 CHRISTENSEN ASSOCIATES
Two-Part RTP Bill
Customer’s bills change from their “Standard” Bill only when they change their hourly loads from the “Baseline” load shape
RTPBill
=StandardBill
LoadHour
+ M.C.Hour
x
September 2002 18 CHRISTENSEN ASSOCIATES
MWh
1 24
Actual
load
Customer “sells” load at high RTP prices
Customer “buys” load at low RTP prices
CBL
Hour of Day
Example of Incremental Energy Charges (Relative to Baseline)
September 2002 19 CHRISTENSEN ASSOCIATES
Economy
Weather
Fuel Price
Unit Availability
Tie Lines
Wholesale Market
RTP RTP PricesPrices
Factors with Major Influence on RTP Prices
September 2002 20 CHRISTENSEN ASSOCIATES
Typical RTP-DA Prices
*Hour at end of interval
HR* cent/kWh status
17 89.4200 A
18 49.5094 A
19 29.2998 A
20 8.2002 A
21 7.6772 A
22 5.2903 A
23 3.6407 A
24 3.2380 A
HR* cent/kWh status
01 3.1440 A
02 3.1151 A
03 2.9661 A
04 2.9329 A
05 2.9307 A
06 2.9384 A
07 2.9980 A
08 3.0449 A
HR* cent/kWh status
09 3.0531 A
10 3.6141 A
11 4.7617 A
12 5.2418 A
13 7.8890 A
14 39.1817 A
15 79.3005 A
16 109.7100 A
September 2002 21 CHRISTENSEN ASSOCIATES
Portion of customers found to respond significantly to RTP prices: 60-75%
Range of flexibility parameters: .01 - .40 (Approximately equal to negative of own-price elasticity)
A short-period price spike of 10 to 20 times the typical price can yield load reductions of 10 to 20% (e.g., 150 MW from 1,000 MW of load)
Do Customers Respond to RTP?Summary of Findings
September 2002 22 CHRISTENSEN ASSOCIATES
Typical Load Response
Increased Usage in All Hours
0:00 4:00 8:00 12:00 16:00 20:00 24:005
6
7
8
9
10
5
6
7
8
9
10
No Response in Low-Price Hours and Decreased Usage in High-Price Hours
(Peak Shavings)
0:00 4:00 8:00 12:00 16:00 20:00 24:00
Increased Usage In Low-Priced Hours and No Response in High-Price Hours(Hiding behind the Baseline Load)
0:00 4:00 8:00 12:00 16:00 20:00 24:00
5
6
7
8
9
10
Increased Usage in Low-Price HoursDecreased Usage in High-Price Hours
0:00 4:00 8:00 12:00 16:00 20:00 24:0056
7
8
9
10
September 2002 23 CHRISTENSEN ASSOCIATES
Demand Profile
Tuesday
0
5000
10000
15000
20000
25000
30000
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24
KW
DE
MA
ND
0
2
4
6
8
10
12
CE
NT
S/K
WH
Cents/kWh Actual kW CBL kW
September 2002 24 CHRISTENSEN ASSOCIATES
Thursday
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 240
20,000
40,000
60,000
80,000
100,000
120,000
140,000
160,000
KW
DE
MA
ND
CE
NT
S/K
WH
Actual kWCBL kWCents/kWh
Demand Profile
September 2002 25 CHRISTENSEN ASSOCIATES
Distribution of RTP Price Elasticities SIC 20 Food Products
0.000
0.050
0.100
0.150
0.200
0.250
0.300
0.350
0.400
0.450
0.500
1 3 5 7 9 11
13
15
17
19
21
23
25
27
29
31
33
35
37
39
41
43
45
47
49
51
53
55
57
September 2002 26 CHRISTENSEN ASSOCIATES
Distribution of RTP Price Elasticities Commercial Office Buildings
0.000
0.050
0.100
0.150
0.200
0.250
0.300
0.350
0.400
0.450
0.5001 3 5 7 9 11
13
15
17
19
21
23
25
27
29
31
33
35
37
39
41
43
45
47
49
51
September 2002 27 CHRISTENSEN ASSOCIATES
Distribution of RTP Price Elasticities Schools and Universities
0.000
0.050
0.100
0.150
0.200
0.250
0.300
0.350
0.400
0.450
0.5001 4 7 10
13
16
19
22
25
28
31
34
37
40
43
46
49
52
55
58
61
64
67
70
73
76
79
September 2002 28 CHRISTENSEN ASSOCIATES
Distribution of RTP Price Elasticities Supermarkets
0.000
0.050
0.100
0.150
0.200
0.250
0.300
0.350
0.400
0.450
0.5001 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35 37 39 41 43 45 47 49 51 53 55 57 59
September 2002 29 CHRISTENSEN ASSOCIATES
0.60
0.70
0.80
0.90
1.00
1.10
1.20
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24
Hour
Nor
mal
ized
Loa
d
0
2
4
6
8
10
12
14
16
Loga
rithm
of
Pric
e (
$/M
W)
Load at highest prices
Highest prices
Reference Load
Load at moderate prices
Reference prices
Moderate prices
RTP-DA Prices and Load Response, by Price Day-type
September 2002 30 CHRISTENSEN ASSOCIATES
Price/Load Response
5
57
414
323
4242
132
92
Days
0
50
100
150
200
250
300
350
400
450
Loa
d R
esp
onse
(M
W)
RTP Price Load Response
RTP-DARTP-DA
September 2002 31 CHRISTENSEN ASSOCIATES
0.00
0.20
0.40
0.60
0.80
1.00
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24
Hour
No
rma
lize
d L
oa
d
0
4
8
12
16
20
Lo
gar
ith
m o
f P
(in
$/M
Wh
)
Load at highest prices
Highest prices
Reference Load
Load at moderate prices
Reference prices
Moderate prices
RTP-HA Prices and Load Response, by Price Day-type
September 2002 32 CHRISTENSEN ASSOCIATES
Price/Load Response
RTP-HARTP-HA
189168
190
319
450
232
Days
0
100
200
300
400
500
Loa
d R
esp
onse
(M
W)
RTP Price Load Response
September 2002 33 CHRISTENSEN ASSOCIATES
Predicted Load Curve Impact
Hour
Hour
mW
0200400600800
100012001400
1 3 5 7 9 11 13 15 17 19 21 23
Original
New
September 2002 34 CHRISTENSEN ASSOCIATES
Load Changes in All Hours
Price Change Hour 17 Only
Original Load
New Load
750
800
850
900
950
1000
1 3 5 7 9 11 13 15 17 19 21 23
September 2002 35 CHRISTENSEN ASSOCIATES
Look at Price Responsive Customers
Average Load
Price Response
497 mW
347 mWNote: Price Response on High Price Days for 24 Customers
$0.20
$0.047
Load Drops 150 mW Morning to Afternoon of High Priced Day
250
300
350
400
450
500
550
1 3 5 7 9 11 13 15 17 19 21 23
September 2002 36 CHRISTENSEN ASSOCIATES
Implications for Demand-Side Strategies
market prices sensitive to demand at high demand levels (elasticity 12)
$70, Retail
Retail Demand
MWs
$400
$200
Supply
Demand response – e.g., RTP – an essential market feature
September 2002 37 CHRISTENSEN ASSOCIATES
Expected Load Changes
(Application load response model to California)*
Price ($/kWh) % Load Change$0.25 2.0%$0.35 3.7%$0.45 5.1%$0.55 6.4%$0.65 7.6%$0.75 8.6%$0.85 9.6%$0.95 10.5%$1.05 11.3%$1.15 12.0%$1.25 12.8%$1.35 13.4%$1.45 14.1%
Note: The reference price is $.15/kWh
Expected Load Changes at Various RTP Price Levels (Percent change)
September 2002 38 CHRISTENSEN ASSOCIATES
What Customers Like About RTP
Access to competitive market prices Low expected energy cost Certainty of cost of consumption changes
September 2002 39 CHRISTENSEN ASSOCIATES
What Do Customers Think?
This chemical company’s threshold is around the 6¢/kWh range. At times they will “buy through” some higher priced hours when they have to meet a customer’s order. They love the rate.
This mining company responded heavily to pricing in July. When the price were over 20¢/kWh, they would curtail below their threshold by a couple of mW.
September 2002 40 CHRISTENSEN ASSOCIATES
What Do Customers Think?
This wood product company break point is around 3.0 cents per kWh. At 3.0 cents or lower, they maximize RTP purchases. At 8.0 cents per kWh, they maximize self generation.
This aggressive commercial facilities manager cuts back florescent fixtures by 1/3, adjusts thermostat, reduces chillers in the afternoon, turns down water heating, and allows the temperature to float on chilled water loop.
September 2002 41 CHRISTENSEN ASSOCIATES
Why Offer RTP?
Lowest priced product based upon sound risk principles
Provides a connection between wholesale and retail energy markets
— Retail prices reflect wholesale costs— Demand response to high prices provides
needed capacity relief and reduces wholesale prices
— Mitigates market power
September 2002 42 CHRISTENSEN ASSOCIATES
Why Two-Part RTP Works
Fundamentally sound market design. Two-part RTP has the same basic structure common to all other commodity markets (future/forward contracts and settlement at spot)
Incremental/decremental load priced at RTP induces efficient consumption/curtailment
The CBL priced at standard tariff provides customer a hedge against price risk without sacrificing efficiency
Voluntary and simple
September 2002 43 CHRISTENSEN ASSOCIATES
Wholesale Markets Employ Vehicles to Mitigate Risks
Few can tolerate the level of price risk; forwards and options essential to electric markets
PRIC
ES
DAYS
$140
$ 30
$ 18
SPOT
FORWARD
September 2002 44 CHRISTENSEN ASSOCIATES
Price Protection Products
Allows RTP customer to manage RTP price risk and volatility
Financial hedge contracts that lock in a price for a specific time period
September 2002 45 CHRISTENSEN ASSOCIATES
CAPS, Collars, Indexes, and Contracts for Differences (CfD’s)
Customer still benefits by reducing load in response to high RTP prices in specific hours
Price Protection Products
September 2002 46 CHRISTENSEN ASSOCIATES
Time Period July, All Hours
Contracted kW 1,000
# Hours 744
Contracted kWh 744,000
CfD Price 6.0 Cents/kWh
CfD Example:
September 2002 47 CHRISTENSEN ASSOCIATES
Settlement for High Average Price: = (10.0-6.0) * 744,000 = $29,760. GPC pays Customer $29,760.
-OR-
Settlement for Low Average Price: = (4.0-6.0) * 744,000 = ($14,880). Customer pays GPC $14,880.
CfD Example:
September 2002 48 CHRISTENSEN ASSOCIATES
Real Time Pricing Summary Benefits Coupled with PPP
Participants— Provides industrial and commercial customers the
cost-based pricing they want and gives them more control over their bill and lower unit costs
Non-participants— Protects them from revenue erosion and benefits
them by allowing Utility to operate more efficiently at a lower cost to all ratepayers
September 2002 49 CHRISTENSEN ASSOCIATES
Utility Company— Provides more efficient pricing and a more
competitive position State
— Attracts new business and rewards business expansion, resulting in increased employment and tax revenues
Real Time Pricing Summary Benefits Coupled with PPP
September 2002 50 CHRISTENSEN ASSOCIATES
Portfolio Pricing on Risk Principles:Expected Customer Cost vs. Price Risk Profile
Risk to Buyer
ExpectedCost
toBuyer
FIP
RTP-DA
VIP RTP w/ Adjustable CBL
RTP w/ PPP
RTP-HA
PLL