Alternative Models of

Electric Industry Restructuring

Bruce Edelston
Director, Southern Company
Presentation to
Carnegie Mellon Electric Industry Center
September 23, 2004
Who We Are
 Southern Company is an investor owned energy company in the Southeastern U.S. and
a holding company for:
Alabama Power Company
Georgia Power Company
Gulf Power Company
Mississippi Power Company
Savannah Electric & Power Company
Southern Power Company
supplying electric service in the states of Alabama, Florida, Georgia, Mississippi.
 Other Businesses
Southern Company Gas
Southern Nuclear
Southern LINC
Southern Telecom
Southern Company Profile
 Generated 183 billion KWh of electricity in
2002 with 39,000 MW
 Earnings for 2002 of $1.3 billion on total
revenues of $10.5 billion
 More than 26,000 employees
 Fortune magazine’s most admired electric
and gas utility in America for the past two
years
 Rates 20% below national average
Generating Mix
 281 generating units at 69 plants in the Southeast
 2004 Generation Fuel Mix:

Gas Oil Nuclear

9.6% 0.1% 15.1%
Hydro
3.7%

Coal
71.5%
ALTERNATIVE MODELS OF
ELECTRIC INDUSTRY
COMPETITION AND
RESTRUCTURING
The Original California Model
Generation --
Utility, IPPs, Marketers Power Exchange/Pool
Contracts
Contracts

Network Operator (ISO)

Contracts

Competitive Suppliers
Distribution Utilities

Contracts
Customers
The Original California Model (cont.)
 All utility generation had to be sold into Power
Exchange
 Customers could buy from the distribution utility,
directly from a generator, or from a competitive
supplier
 Distribution utility had to buy all its needs from
the Power Exchange
 Retail rates of distribution utilities remained
regulated
 Competitive suppliers could buy their needs from
generators or from the Exchange, or some
combination
Customer Choice Model (TX)
Generation --
Competitive Suppliers
Utility, IPPs, Marketers

Network Operator (RTO/ISO)

Contracts Contracts

T & D Utilities

Customers
Customer Choice Model (cont.)
 Customers may contract directly with
generators or competitive suppliers
(power marketers) for their own needs
 Network operator runs transmission
system, does planning and scheduling,
balances supply and demand through bid-
in balancing market, and is responsible
for reliability
 Distribution company simply operates
distribution system - it may put out bids
for “standard offer service”
Centralized Dispatch Model (PJM)

Generation --
Utility, IPPs, Marketers

Centralized Pool (e.g., PJM)

Network Operator

Competitive Suppliers
Distribution
- Billing
- “POLR” Service
- Value Added Services
Centralized Dispatch Model (cont.)
 Retailers (either utilities or competitive
suppliers) buy all of their needs from pool,
resell to end users
 All generators bid into pool on an hourly
basis
 Pool dispatches generation from lowest
cost bid to highest cost bid
 Highest cost bid that gets dispatched
becomes market clearing or “spot” price
 All generators that are dispatched are
paid the spot price
Centralized Dispatch Model (cont.)
 Pool is either “energy only” or
energy and capacity are separate
products (and markets)
 May be a minimum capacity
requirement for suppliers
 Customer choice is really a matter of
risk management for suppliers
Vertically-Integrated, Incremental
Wholesale Competition Model
Existing Generation -- New Generation –
Regulated Competitive

Network Operator (RTO/ISO)

Integrated Utility
(Plans and Operates)

Distribution
Existing Needs - Use Own Units
Incremental Needs -- Buy from Market
Vertically-Integrated, Incremental
Wholesale Competition Model (cont.)
 No customer choice (limited exceptions)
 Existing generation used for retail sales remains
regulated
 New generation and existing (excess) generation
available for wholesale sales are market-based
(assuming regulatory approval)
 Integrated utilities with service obligations buy
incremental needs via requests for proposals
 Utilities may or may not bid a “self-build” (rate
base) option
 Utilities choose incremental option based on
price and non-price factors and signs purchase
power agreement (typically 5-7 years)
Vertically-Integrated, Incremental
Wholesale Competition Model (cont.)
 Utility affiliates may also bid if permitted
by state regulators
 Over time, more and more generation is
acquired through purchase power
agreements, rate base diminishes
 Transmission and distribution planning
and operations continue to be performed
by integrated utility
 Integrated utility also distributes power
and makes retail sales at rates set by
state regulators
CURRENT
MODELS
Customer Choice States

Retail Choice State

Non-Retail Choice State
Source: EIA
Centralized Dispatch Areas

 New England
 New York
 PJM
 ERCOT
 California
 Planned: Midwest ISO
Vertically-Integrated States

Unbundled
Vertically-Integrated
Source: EIA
Comparison of Models
Customer Choice - Pros

 Lets customers decide

 Risks shifted from customers to
suppliers
 Spurs product innovation
 Spurs technical innovation
 Maintains competitiveness of
economy
Customer Choice: Issues
Are Prerequisites for Competition Satisfied?

Competitive Markets: Electric Markets:

Consumers should face true
costs of supply
Customers should be
responsive to prices
Large number of sellers and
buyers
Customers can choose level of
reliability
No inter- and intra-class
subsidies
Consumers have regulated
option available, regulators likely
to cap very high prices
Demand is very inelastic
Few sellers, few buyers
Reliability is a public good
Subsidies abound
Customer Choice – Issues (cont.)
 Do small customers want choice?
 Can regulators/politicians let competition
work?
 Who will pay for reserves needed for
reliability but not revenue producing?
 Are there significant economies of scope
that are lost by unbundling?
 Should any one care about fuel diversity?
 What about externalities?
 Transaction costs vs. savings?
 How is success measured?
Centralized Economic Dispatch:
Pros
 Production efficiency clearly the greatest
potential benefit
 Locational price signals tell where generation
and transmission should be built
 Physical system is separated from financial
system
 Relatively easy for areas with existing traditional
power pools
 Provides hourly price signals to customers
 Lessens market power concerns
 PJM has made it work (some experience)
Centralized Economic Dispatch:
Issues
 Costs of RTOs (vs. benefits)
RTO Costs (2003)
Revenue Cost per Unit
Requirement ($/MWh)
PJM $252,164,806 0.723
NYISO 117,578,796 0.718
ISO–NE 102,924,000 0.787
CA ISO 235,240,000 1.020
ERCOT 184,159,748 0.545
Ontario 107,204,400 0.705

Source: Public Power Council

 RTO Cost Trends (2000-2004)

1.2

1
All-In Cost ($/MWh)

0.8

0.6

0.4

0.2

0
2000 2001 2002 2003 2004

PJM NYISO NE-ISO CA ISO ERCOT Ontario

Source: Public Power Council

Centralized Economic Dispatch:
Issues (cont.)
 Will regulators (politicians) accept
price volatility and high prices
necessary to pay for peakers (that
must recover fixed costs in only a
few hours per year)?
 If not, how will sufficient generating
capacity be ensured?
Centralized Economic Dispatch:
Issues (cont.)
 How will demand side interact with
pool?
 Who will build transmission? What
incentives will they have?
 Fuel diversity
 Externalities/public benefits
Centralized Economic Dispatch:
Issues (cont.)
 What is appropriate size?
 Seams issues
 Requires transfer of jurisdiction from
states to FERC
 Reliability responsibilities dispersed
(more complicated)
 Unregulated utilities (coops and
government-owned) must participate,
especially where they are a major
presence (NW, SE)
Vertically-Integrated, Incremental
Wholesale Competition: Pros
 Clear accountability for reliability
and service obligations
 Fuel choice and externalities can
remain part of resource planning
 Generation, transmission and
distribution can be planned jointly,
lowering total costs (economies of
scope)
Vertically-Integrated, Incremental
Wholesale Competition: Pros (cont.)
 Because of stranded cost recovery, most
of the benefits of competition come from
incremental generation, not existing
 Customers get benefits of wholesale
competition without transaction costs
(and hassle) of choosing supplier
 Integrated utility manages risks on behalf
of customers
 States retain jurisdiction
Vertically-Integrated, Incremental
Wholesale Competition: Issues
 Perception of market power
– Generation dominance
– Transmission access
– Other barriers to entry
 No transparency of dispatch
 Few buyers
 Lack of regional planning
 Retail customers retain risks of bad
utility decisions
Possible Additions to Vertically-
Integrated Model
 Independent operation of OASIS and
granting of interconnections and
transmission access
 Regional planning and security
coordination by independent entity
 Short-term formal competitive
procurement process
 More formalized long-term RFP process
with greater transparency and
independent oversight
Other Critical Issues
 Lack of investment in both generation and
transmission
 Credit ratings of IPPs/marketers
 Relationship between federal and state
regulators
 August 14 blackout and its ramifications
 Lack of mandatory reliability rules
 Tug of war between environmental
objectives and competitive objectives
 Elected vs. appointed Commissions
Conclusions
 All of these models (except
California) can work if issues are
properly addressed
 Regional characteristics and
concerns drive choices
 Competition should be a means to
an end (reliability at lowest possible
cost) rather than the end itself