Transcript Ensuring Generation Adequacy Through Hedging Obligations

Ensuring Generation Adequacy
Through Hedging Obligations
Shmuel Oren
University of California at
Berkeley
NSF/EPRI Workshop on Economics, Electric
Power and Adaptive Systems
Washington DC, March 28-29,2002
What is Reliability
 NERC (North American Electric Reliability Council) defines reliability
as: “the degree to which the performance of the elements of the
electrical system results in power being delivered to consumers
within accepted standards and in the amount desired”
 Reliability encompasses two concepts:
 Security: “the ability of the system to withstand sudden disturbances.”
This aspect concerns short-term operations and is addressed by
ancillary services which include:Voltage support, Congestion relief,
Regulation (AGC) capacity, Spinning reserves, Nonspinning reserves,
Replacement reserves.
 Adequacy: “the ability of the system to supply the aggregate electric
power and energy requirements of the consumers at all times”. This
aspect concerns planning and investment and is addressed by Planning
reserves, Installed capacity, Operable capacity or Available capacity.
Shmuel Oren – March 28-29, 2002
Planning Reserves and Reliability
 Plentiful reserve capacity makes it easier to achieve security but is not
necessary
 Security can be achieved even with limited reserves by curtailing load or
by raising prices sufficiently to induce demand response and investment.
 Generation adequacy is required to insure supply at a reasonable price
 In a competitive market “obligation to serve” is “obligation to serve at a
price”
 “Reliability” is not a product and has no meaning unless it is defined in
terms of a price ceiling on secure energy supply
 Generation adequacy can be interpreted as price insurance which in
theory is a private good. Given proper technology, market participant
should decide how much of the good they want and at what price,
based on their risk management preferences.
Shmuel Oren – March 28-29, 2002
Achieving generation
adequacy in an ideal market
 Inadequate supply leads to high prices which attract investment
 Excess capacity will drive competitive prices to marginal cost
 Generators on the margin and reserve capacity will not cover their
fixed costs.
 Exit through early retirement of capacity will drive prices up during
peak demand leading to demand response and scarcity rents
 When capacity is optimal scarcity rents exactly covers fixed costs
 Capacity deficiency will drive scarcity rents above equilibrium levels
resulting in excess profits which will attract new investment.
 Forward markets will form that enable consumers to lock in prices
and avoid price volatility and to investor to hedge their investment
risk by securing long term supply contracts
Shmuel Oren – March 28-29, 2002
Energy Market With Excess Capacity
Energy
Price
($/MWh)
Demand at 7:00 - 8:00 p.m.
Price at7:00-8:00 p.m
Demand at
9:00 - 10:00 a.m.
Price at
9:00 10:00 a.m.
Demand at
2:00 - 3:00 a.m.
Price at
2:00 3:00 a.m.
GEN 1
GEN 2
Q1
GEN 3
GEN 4
Q2
Shmuel Oren – March 28-29, 2002
GEN 5
GEN 6
Optimal
Capacity
Operating MW
level
Energy Market With Optimal Capacity
Energy
Price
($/MWh)
Demand at 7:00 - 8:00 p.m.
Price at7:00-8:00 p.m.
Scarcity rent
Price at7:00-8:00 p.m. with Capacity payment
Demand at
9:00 - 10:00 a.m.
Price at
9:00 10:00 a.m.
Demand
Response
Demand at
2:00 - 3:00 a.m.
Price at
2:00 3:00 a.m.
GEN 1
GEN 2
Q1
GEN 3
GEN 4
Q2
Shmuel Oren – March 28-29, 2002
GEN 5
Optimal
Capacity
MW
Energy price volatility
 Price volatility is an inherent aspect of electricity due
to its nonstorability and the steep supply curve.
WSCC Generation Resource Stack
$160.0
Electricity
On-peak
Spot Prices
Electricity On-Peak
Spot Price
Spot Price
ERCOT on-peak
$140.0
$100 Marginal Cost
($/MWh)
$90
Marginal Cost
Resource stack includes:
Hydro units;
Nuclear plants;
Coal units;
Natural gas units;
Misc.
$80
$70
$60
$50
COB on-peak
$80.00
$70.00
$120.0
$60.00
$100.0
$50.00
$80.0
$40.00
$60.0
$30.00
$40.0
$20.00
$20.0
$10.00
$40
$30
$20
$10
$0
0
20000
40000
60000
80000
100000
$0.0
12/1/199 3/10/199 6/18/199 9/26/199 1/4/1997 4/14/199 7/23/199 10/31/19 2/8/1998 5/19/199
5
6
6
6
7
7
97
8
Demand (MW)
Time
Shmuel Oren – March 28-29, 2002
$0.00
Key questions
 Can we rely on the “market” to provide investment
incentives for adequate planning reserves?
 What mechanism will provide an income stream that can
sustain reserve generation capacity?
 Will capital markets operate efficiently to sustain an
adequate amount of generation investment?
 Is an unrestricted energy market in which scarcity rents
feed new investment politically feasible?
 What mechanism should be used (if any) to restrain market
power and transfer of wealth between producers and
consumers while investment catches up with scarcity?
 What should be the risk management obligation of an LSE
Shmuel Oren – March 28-29, 2002
What's missing in the
theoretical market paradigm?
 Theory does not account for reserve capacity that is required on the “bench” to
ensure system reliability. When resources are only paid for produced energy such
reserve capacity will not collect sufficient revenues to cover its fixed costs and
will exit the market
 Steep supply function and uncertainties make scarcity rent highly volatile and
sensitive to market error in determining the optimal capacity
 It is impossible to differentiate legitimate scarcity rents from inflated prices due
to exercise of market power.
 Demand response is limited by technological barriers and operational practices
 Very high scarcity rents even if they are legitimate are politically unacceptable
(reason for price caps)
 Low levels of reserves foster collusive behavior and market power
 Even suppliers with low market share can become pivotal suppliers.
 High prices are sticky
 Capacity shortages cannot be resolved overnight and while the entry occurs the
persistent scarcity rents result in wealth transfers from consumers to producers.
 Exposures in the electricity supply chain are not properly allocated to insure
voluntary, socially efficient risk management practices by the market participants
Shmuel Oren – March 28-29, 2002
Alternative Approaches to Ensuring
Generation Adequacy
 Capacity payments (old UK system, Argentina, Spain)
Generators receive capacity payments based on availability,
technology, VOLL, LOLP to incentivize investment and availability.
 Shortcomings:
 Payoff to incumbents but does not reassure investors
 Results in over investment and too low energy prices that reinforce the
need for capacity payments
 Suppresses demand side response since scarcity rents are covered by
capacity payments
Shmuel Oren – March 28-29, 2002
Alternative Approaches to Ensuring
Generation Adequacy (cont’d)
 ICAP obligation (PJM, New York, New England)
Central agency (ISO or Regulator) specifies requirements for
planning reserves based on traditional planning tools.
Load serving entities have to meet a monthly prorata ICAP
obligation
ICAP markets allow supplier to trade reserves and efficiently
reallocate the reserves requirements.
 Shortcomings:
 Capacity product is too short term to affect planning
 Capacity product does not obligate the seller to be available for energy
production or to provide energy at some price
 Capacity prices do not reflect the value of producing energy
 Short-term supply and demand for ICAP are inelastic so there is either
excess (zero price) or shortage (infinite price)
Shmuel Oren – March 28-29, 2002
Alternative Approaches to Ensuring
Generation Adequacy (cont’d)
 ACAP obligation (Proposed in California)
ISO specifies requirements for available capacity obligation
Load serving entities have to meet a monthly ACAP obligation that
is based on their forecasted next month peak load plus a fixed
percentage
ACAP obligation can be met through a portfolio of generation
resources an physical load management
Resources counted toward ACAP obligation are subject to ISO
verification at the beginning of each month and must be scheduled
or offered into the ISO operated markets.
 Shortcomings:
 Based on the outdated “obligation to serve” paradigm were capacity is a
product and reliability is a service attribute.
 In a market paradigm, capacity is an option to produce energy and
reliability is the availability of supply at a reasonable price.
 Without an explicit strike price a capacity product does not protect
customers from price spikes and therefore does not provide reliability
Shmuel Oren – March 28-29, 2002
Alternative Approaches to Ensuring
Generation Adequacy (cont’d)
 Two year RPRS obligation (Reliant proposal)
Works like another ancillary service product
ISO procures two year options on RPRS capacity based on
forecasted need of replacement reserves
QSEs assigned obligation based on daily load and charged RPRS
clearing price for their obligation
Providers required to offer contracted capacity as replacement
reserves at contract strike price and offer balancing energy at MCPE.
Contract duration adequate to affect planning
 Shortcomings:
 Cost of option not covered by QSE payment. Must be uplifted or allocated in
proportion to annual RPRS payment.
 Puts ISO in the position of buying forward and selling spot
 Inconsistent with decentralized markets and MIN ISO philosophy
Shmuel Oren – March 28-29, 2002
Alternative Approaches to Ensuring
Generation Adequacy (cont’d)
 Hedging obligation (strawman)
 load serving entities (LSEs) are required to hold at the beginning of each month
verifiable hedges in the form of forward contracts and/or call options totaling
115% of their next month forecasted peak load.
 Qualifying hedges must have at least two years duration with no less than one
year remaining life. Strike prices of call option should be at or below a maximum
level set by the regulator (the strike price should be substantially below the price
cap, e.g. 200$/MWh)
 Hedging obligations can be met by a portfolio of contracts with generators and
curtailable load contracts (physical cover).
 Generator risk (and consequently the cost of options) may be reduced by
indexing the strike price to fuel cost or by using “spark spread” call options
(spark spread=electricity price - heat rate adjusted gas price)
 Call options that are exercised must be able to generate the promised power or
be liable for the price cap applied to the undelivered quantity.
 Self-insurance covered by a financial security may be allowed on a limited basis
If allowed, security should cover the difference between the price cap and the
regulated strike price for an X month worth of the uncovered hedging obligation.
 ACAP can be viewed as special case were the strike price equals to the price cap
and physical exercise capability is enforced
Shmuel Oren – March 28-29, 2002
Key aspects of hedging obligations
 Emphasis on mitigating price volatility rather than on “steel in the
ground” which is only one of the possible market responses.
 Multiple means of meeting hedging obligation ensures balance
between investment, demand response and risk management
 Hedging products are long term to facilitate new investment response
by transferring risk from the investor to the LSE.
 Enables reserve generation capacity to secure a stable income stream
for fixed cost recovery in exchange for a tangible obligation to
produce energy at a reasonable price when needed.
 LSE obligations revised monthly to reflect changes in customer base
 Secondary market for call options will enable LSE to adjust their
holdings. Prices will fluctuate according to market conditions. (e.g.
daily fluctuation of long term treasury bond prices)
Shmuel Oren – March 28-29, 2002
Summary
 Generation adequacy should be viewed and treated as a financial risk
management issue rather than a reliability issue.
 The ultimate goal of long term reserves policies in a competitive
electricity market is to mitigate price volatility through alternative means
and not just to promote “steel in the ground”
 Imposing hedging requirements on LSEs is a market friendly way of
ensuring generation adequacy
 While price caps and reliability standards should be subject to regional
jurisdictions (RTO, FERC, NERC) hedging requirements imposed on LSEs
should be regulated at the state level.
Shmuel Oren – March 28-29, 2002
Power Economics at UC
Berkeley
Econ./Bus Ad./(UCEI):
Bornstein, Bushnell, Spiller, Wolfram
Emphasis on empirical IO, market power analysis,
environmental issues, political economy
EECS:
Wu, Variya
Emphasis on market design/operations,
communications, control
IEOR
Shmuel Oren
Emphasis on market design, planning, scheduling,
risk management, auctions
Shmuel Oren – March 28-29, 2002
Power Systems Theses
(at least one comm. member from Econ, Bus Ad and/or EECS)
Joseph Doucet "Differential Pricing of Electricity Through Interruption Insurance",
1988
Todd Strauss, "Interruptible Electricity Tariffs with Early Notification", 1992
Alva Svoboda "Simulation of Dispatchable Demand-side Management in Electric
Power System Operation Planning", 1992
Eric Friedman "Topics in Coordination and Decentralization", 1993
James Bushnell, "Multi-Dimensional Revelation in Auctions for Electric Power
Supply", 1993
Chung-Li Tseng, "On Power System Generation Unit Commitment Problems", 1996
Wedad Elmaghraby "Multi-unit Auctions for Electric Power with Nonconvex Costs",
1998
Shijie Deng "Financial Methods in Deregulated Electricity Markets", 1999
Ami Beth Craft "Market Structure and Capacity Expansion in an Unbundled Electric
Power Industry" (Stanford MSE Dept.), 1999
Rajnish Kamat, “Market Mechanisms in Deregulated Electricity Markets”, 2001
Afzal Siddiqi “Equilibrium Analysis of Spot and Forward Markets for Energy and
Reserves” 2002
Shmuel Oren – March 28-29, 2002
Typical coursework
 Major in IEOR
 3-4 courses in optimization (LP,NLP, IP, Networks,Combinatorial, DP,)
 2-3 course in stochastic modeling and optimization (Markov processes,
queuing, SDP)
 Minor or equivalent in Econ
 Micro
 Game theory
 Mechanism design
 Econometrics
 Industrial organization
 Minor or equivalent in Finance
 Corporate finance
 Derivatives
 Dynamic asset pricing
Shmuel Oren – March 28-29, 2002
Key Points
 Economic work requires economic thinking. Engineering
students doing economics work should learn to focus on
process analysis rather than the problem/answer
paradigm.
 Encourage students to acquire broad training and
exposure to other disciplines.
 Do not reinvent the wheel. Encourage students to
search the literature in other disciplines and build on it.
 Encourage students to publish their work in the
appropriate disciplinary journals. Avoid the “cottage
industry” syndrome.
Shmuel Oren – March 28-29, 2002